Technical analysis connects price structure, participation, investor behavior, and quantitative evidence. This guide develops those skills through 20 original multiple-choice questions and 60 substantive concepts, including chart construction, Dow and Wyckoff principles, momentum, volatility, systematic testing, comparative performance, risk, and ethics. It follows the supplied 2026 CMT Level I scope for October 16, 2026: Theory and History 38%, Classical Techniques 33%, Advanced Techniques 26%, and Ethics 3%. The preview allocates questions 8/6/5/1 and concepts 23/20/15/2; these are practice allocations. Calculations state their assumptions, and examples distinguish conditional evidence from guarantees. Source s3 establishes syllabus relevance; technical explanations use established subject knowledge.
What this guide covers
Explore 20 original CMT Level I questions and 60 technical-analysis concepts, with worked examples and clear rationales.
Theory and History
1. Dow, Hamilton, and Rhea: from observations to theory
Charles Dow's market editorials supplied observations that later writers, including William Peter Hamilton and Robert Rhea, organized into what became known as Dow Theory. Its historical importance lies in treating market averages, trends, and confirmation as systematic evidence rather than isolated tips. Dow did not publish a modern mechanical trading system with universally fixed parameters. Distinguish the origin of the ideas from their later codification. This history explains why contemporary summaries share central principles while sometimes differing in operational definitions and reversal rules.
Worked example: A researcher finds two textbooks using different numerical filters for a Dow Theory reversal. She separates the shared principle of confirming important trend changes from each author's implementation, rather than assuming Dow prescribed one universal percentage filter.
Common mistake: Attributing every later mechanical interpretation to Dow himself confuses the historical foundation with subsequent authors' rules and refinements.
Official syllabus reference: [1]
2. Market action discounts information and expectations
The discounting premise treats price as the aggregate outcome of participants acting on information, expectations, constraints, and preferences. Technical analysis studies that observable outcome without requiring a complete explanation of every underlying cause. A favorable announcement can accompany a decline if expectations were higher or other information dominates. Discounting is an organizing assumption, not proof that prices equal intrinsic value or that all participants possess identical knowledge. Apply it by interpreting the response to an event alongside the event itself, rather than assigning direction from headlines alone.
Worked example: A retailer reports rising earnings, yet its shares break support after management lowers its outlook. The analyst recognizes that the observed decline can incorporate forward expectations even though the backward-looking earnings number appears favorable in isolation.
Common mistake: Assuming good news must raise prices overlooks expectations already embedded in the market and information accompanying the headline.
Official syllabus reference: [1]
3. Primary, secondary, and minor market movements
Dow Theory distinguishes a broad primary trend from secondary reactions and shorter minor fluctuations. These movements coexist: a substantial pullback can occur within a continuing primary advance. Classification depends on the scale and significance of the movement, rather than a universal number of bars. The distinction helps prevent a short-horizon reversal from being mistaken for a change in the larger market direction. Identify the analytical horizon first, then compare meaningful peaks and troughs at that scale while recognizing that smaller countertrend movements remain possible.
Worked example: A weekly chart shows successive higher major lows over a year, while daily prices decline for several weeks. An analyst can describe the daily decline as a secondary reaction without declaring that the primary uptrend has already reversed.
Common mistake: Calling every short decline a bear market collapses different trend scales and obscures the evidence needed for a primary reversal.
Official syllabus reference: [1]
4. Confirmation between market averages
Dow Theory seeks agreement between related market averages when assessing important directional developments. Historically, industrial and transportation averages represented complementary aspects of economic activity. A significant new high in only one average leaves the advance unconfirmed; corresponding strength in both provides broader evidence. Confirmation need not occur simultaneously, and nonconfirmation does not automatically establish the opposite trend. The practical decision is whether the evidence supports a confirmed development, an unresolved disagreement, or a later reversal based on additional price structure.
Worked example: An industrial average exceeds its previous major high on Monday. A transportation average does so two weeks later. The analyst records delayed confirmation rather than rejecting the signal because the two breakouts occurred on different trading days.
Common mistake: Treating nonconfirmation as an immediate sell signal confuses missing supporting evidence with positive evidence of an established bearish reversal.
Official syllabus reference: [1]
5. Volume as a confirming variable in Dow Theory
Volume measures trading activity and can help assess whether participation supports the prevailing price direction. In a healthy advance, stronger activity on upward movements and lighter activity on reactions is traditionally considered supportive. Price remains the primary evidence; volume is a secondary confirmation rather than an independent guarantee. Volume interpretation also depends on the instrument and data source because exchange volume, consolidated equity volume, and decentralized-market proxies are not equivalent. Compare like measurements over comparable intervals before drawing conclusions about participation.
Worked example: An index rises on expanding constituent turnover and retreats on quieter sessions. The analyst treats that pattern as supportive of the advance, while still requiring price evidence to identify the trend and any subsequent reversal.
Common mistake: Equating high volume with bullishness ignores whether trading activity accompanies upward progress, downward pressure, or a contested price range.
Official syllabus reference: [1]
6. Trend persistence and evidence of reversal
A trend is generally treated as continuing until sufficient contrary evidence appears. This principle prevents analysts from calling a turning point merely because a move seems extended. In an uptrend, progressively higher significant highs and lows support persistence; disruption of that structure raises reversal concerns. Different methods define confirmation differently, so the reversal rule must be stated rather than assumed. A broken trendline can indicate a change in slope before the full peak-and-trough structure reverses. Persistence is a working hypothesis, not a claim that trends continue indefinitely.
Worked example: A stock breaks a steep rising trendline but remains above its last important reaction low. The analyst distinguishes a slowing advance from a fully established downtrend and watches whether the next rally and decline change the peak-and-trough sequence.
Common mistake: Predicting reversal solely because a trend is old substitutes a judgment about duration for observable evidence of changed market structure.
Official syllabus reference: [1]
7. Wyckoff accumulation and distribution
Wyckoff analysis interprets trading ranges as possible accumulation or distribution processes before directional movement. Accumulation involves absorption of available supply before markup; distribution involves transfer of holdings before markdown. A sideways range alone does not identify which process is occurring. Analysts examine tests of boundaries, volume, price spread, and the quality of subsequent breakouts or breakdowns. The composite-operator idea is a conceptual lens for organized buying or selling, not evidence that one identifiable actor controls the market. Interpretation remains provisional until subsequent behavior supports it.
Worked example: A stock repeatedly tests a range low with shrinking downward progress, then rallies through resistance on stronger participation. Those observations support an accumulation interpretation more than the range's existence alone would, although a failed breakout would weaken the conclusion.
Common mistake: Labeling every sideways market accumulation ignores distribution, reaccumulation, redistribution, and ranges that provide no clear directional evidence.
Official syllabus reference: [1]
8. Wyckoff's law of supply and demand
The supply-and-demand principle interprets price movement through the relative urgency and available liquidity of buyers and sellers. Every completed transaction has both a buyer and seller, so rising prices do not mean that executed buyers numerically outnumber executed sellers. Prices rise when aggressive buying must transact at progressively higher offers or available supply withdraws. Falling prices reflect the corresponding pressure on bids. Technical analysis infers this balance from observable price and volume behavior rather than directly counting investors who feel bullish or bearish.
Worked example: A large market buy order consumes offers at $30.00, $30.02, and $30.05. Each execution has a seller, but the buyer's urgency against limited offered liquidity moves the transaction price upward across those levels.
Common mistake: Saying prices rise because every trade has more buyers than sellers misunderstands matched transactions and the role of urgency and liquidity.
Official syllabus reference: [1]
9. Wyckoff's cause-and-effect principle
Cause-and-effect reasoning connects a period of preparation, often a trading range, with a later directional movement. In Wyckoff-related point-and-figure work, horizontal counts can be used to estimate an indicative price objective under specified counting conventions. The principle does not mean that elapsed calendar time alone determines the size of the next move. Range structure, count selection, box size, and reversal settings affect the estimate. Apply the idea as a structured relationship between preparation and potential movement, while keeping forecasts conditional on market evidence.
Worked example: Two stocks spend the same number of weeks in ranges, but one develops a much broader point-and-figure count under identical settings. An analyst does not assign equal objectives simply because the calendar durations match; the specified count structure matters.
Common mistake: Converting range duration directly into a guaranteed target replaces a conditional counting method with an unsupported deterministic relationship.
Official syllabus reference: [1]
10. Wyckoff's effort-versus-result comparison
Effort-versus-result analysis compares trading activity with the price progress achieved. Volume commonly represents effort, while price spread, net change, and follow-through represent result. Large volume with little upward progress may indicate substantial selling absorbing demand, but its meaning depends on location and subsequent behavior. Similar compression near a low can reflect buyers absorbing supply. The essential task is identifying whether activity and price response reinforce one another or diverge. Neither high volume nor a narrow bar has a fixed interpretation when isolated from surrounding structure.
Worked example: Near resistance, a stock records its heaviest volume in weeks but closes only slightly higher and fails to advance the next day. The analyst considers whether supply is absorbing aggressive buying rather than treating the volume surge as automatically bullish.
Common mistake: Reading heavy volume as strength without examining the achieved price movement misses the central comparison between trading effort and result.
Official syllabus reference: [1]
11. Trend following versus mean reversion
Trend-following methods assume that an established directional move may persist sufficiently to exploit continuation. Mean-reversion methods assume that a deviation from an equilibrium proxy may subsequently contract. The same price extension can therefore be interpreted differently by the two approaches. Neither premise is universally valid: market regime, horizon, liquidity, and the definition of equilibrium matter. A moving-average distance measures extension but does not determine which process will dominate. Distinguish the mechanism being proposed before selecting indicators or interpreting a breakout and an overbought reading.
Worked example: A stock closes well above its recent average after leaving a long range. A trend follower sees possible continuation, while a mean-reversion trader sees extension. Their disagreement reflects different hypotheses, which require different entry, exit, and failure conditions.
Common mistake: Combining continuation entries with arbitrary mean-reversion exits can create inconsistent rules unless the distinct mechanisms and horizons are explicitly reconciled.
Official syllabus reference: [1]
12. Weak, semi-strong, and strong market efficiency
The efficient market hypothesis distinguishes information sets. Weak-form efficiency concerns past market data; semi-strong efficiency includes publicly available information; strong-form efficiency includes private information as well. Stronger forms contain the weaker information sets. A price-based rule showing reproducible abnormal returns after costs would directly challenge weak-form efficiency, but ordinary positive returns would not suffice. Risk compensation, chance, data mining, and implementation assumptions must be considered. These forms describe hypotheses about information incorporation rather than assertions that prices never change or investors never make mistakes.
Worked example: A historical-price strategy earns 9% while an appropriate risk-matched benchmark earns 10%. Its positive return alone does not challenge weak-form efficiency because the relevant question is abnormal performance after risk and implementation costs, not whether money was made.
Common mistake: Treating any profitable chart signal as a refutation of efficiency overlooks risk exposure, costs, chance, and whether the result was selected afterward.
Official syllabus reference: [1]
13. Random walks, independence, and changing volatility
A simple random-walk model treats successive price increments as unpredictable from prior increments under its assumptions. This differs from claiming that prices cannot trend visually or that risk is constant. Even independent increments can produce apparent runs and chart patterns by chance. Financial returns may also exhibit volatility clustering, where large absolute moves tend to follow large absolute moves without making the next move's sign equally predictable. Separate predictability of direction, expected return, and conditional variance when evaluating evidence against a particular random-process model.
Worked example: After several unusually large daily moves, an analyst forecasts elevated near-term volatility but does not forecast an upward next-day return. That distinction allows predictable risk conditions without assuming the direction of the next price change is predictable.
Common mistake: Assuming volatility clustering proves directional predictability confuses dependence in return magnitude with dependence in the sign or conditional mean.
Official syllabus reference: [1]
14. Fractals and nested market structure
Fractal descriptions emphasize similar structural forms across observation scales. Trends, reactions, and consolidations can appear on intraday, daily, and weekly charts. This does not require exact geometric repetition or identical forecasting performance. A small-scale advance may be only a countertrend movement within a larger decline, so the same instrument can legitimately have different trend labels at different horizons. Apply the idea by stating the chart interval and relevant structure explicitly. Scale resemblance supports a descriptive framework; it does not turn a short-term signal into a guaranteed long-term forecast.
Worked example: A five-minute chart shows higher highs and lows while a weekly chart remains below declining major peaks. The analyst identifies a short-term uptrend nested within a longer downtrend rather than forcing both charts to carry the same label.
Common mistake: Assuming a pattern on one time frame must control another mistakes structural resemblance for a guaranteed causal relationship across horizons.
Official syllabus reference: [1]
15. Adaptive markets and changing strategy effectiveness
The adaptive markets perspective links market efficiency and behavior to competition, learning, and changing environments. A profitable opportunity can attract capital, alter participant behavior, and diminish as it becomes crowded. New institutions, technology, constraints, or shocks can create different opportunities. This perspective does not guarantee that every strategy cycles predictably between success and failure. Its application is to treat a historical edge as conditional on the environment rather than permanent. Changes in performance should be examined against plausible mechanisms, implementation changes, and statistical uncertainty.
Worked example: A short-term reversal strategy weakens after trading costs fall and competing firms enter the same market. The analyst considers whether faster arbitrage has reduced the opportunity instead of assuming the original historical average remains a permanent market property.
Common mistake: Explaining every losing period as adaptation makes the theory unfalsifiable unless the proposed environmental mechanism and supporting evidence are specified.
Official syllabus reference: [1]
16. Contrary opinion and sentiment extremes
Contrary-opinion analysis asks whether unusually one-sided expectations or positioning leave less capacity for additional buying or selling in the prevailing direction. A crowded bullish view can create vulnerability if expectations disappoint, but bullish sentiment often accompanies sustained advances. Survey opinions, options activity, and actual positions measure different phenomena and should not be treated interchangeably. An extreme reading is therefore a condition to interpret, not an automatic reversal instruction. Historical comparison, participant composition, and price response help distinguish a persistent consensus from a potentially exhausted one.
Worked example: A bullish survey reaches an unusually high historical percentile while prices continue breaking higher. The analyst records crowding risk but does not declare an immediate top; a later failed breakout would add different evidence about the market's response.
Common mistake: Selling whenever most respondents are bullish ignores that trends can persist and that expressed opinions do not directly measure remaining buying capacity.
Official syllabus reference: [1]
17. Prospect theory: reference points and loss aversion
Prospect theory evaluates outcomes relative to a reference point rather than solely by final wealth. Its value function commonly describes diminishing sensitivity as gains or losses grow and a steeper response to losses than comparable gains. Loss aversion concerns that asymmetry; it is distinct from risk aversion in every situation. The framework can also produce risk seeking when choices are framed as losses. Application requires identifying the reference point and choice structure. A purchase price, prior peak, or stated target may become psychologically influential without being economically decisive.
Worked example: An investor feels a $500 loss much more intensely than a $500 gain and evaluates both against the account value at the start of the month. The analyst identifies the reference point and asymmetric response without assuming a universal numerical loss-aversion coefficient.
Common mistake: Equating loss aversion with refusal to take any risk misses the possibility of risk seeking when trying to escape a perceived loss.
Official syllabus reference: [1]
18. The disposition effect in realized gains and losses
The disposition effect is a tendency to realize winners relatively readily while retaining losers, often because realizing a loss is psychologically painful. It concerns selling behavior rather than merely remembering a purchase price. Reference dependence, regret avoidance, and mental accounting can contribute. Holding a losing position is not automatically biased if current evidence and objectives justify it. Diagnose the effect by comparing decisions with the forward-looking investment case and by looking for systematic asymmetry between comparable winning and losing positions, rather than labeling every delayed sale irrational.
Worked example: A trader quickly sells modest winners to lock in satisfaction but repeatedly postpones exits from otherwise comparable losing trades. Reviewing the pattern across many decisions reveals an asymmetric realization tendency more clearly than examining one losing position alone.
Common mistake: Calling every retained loser a disposition error ignores whether new information and the current opportunity set independently justify continued ownership.
Official syllabus reference: [1]
19. Anchoring and insufficient adjustment
Anchoring occurs when an initial or salient value exerts excessive influence on subsequent judgment. Analysts may adjust away from a purchase price, old target, or prior market high but fail to adjust enough when relevant evidence changes. The anchor need not be economically meaningful. Distinguish anchoring from a legitimate benchmark: a prior price can be relevant to market structure, yet its relevance must come from observable supply, demand, or the analytical method. Apply fresh valuation or technical evidence before allowing a familiar number to dominate the conclusion.
Worked example: A stock's earnings outlook deteriorates, but an analyst revises a $100 target only to $98 because $100 feels like the natural starting point. An independent estimate based on the revised assumptions would have produced a substantially different value.
Common mistake: Assuming a familiar historical price is fair value confuses psychological salience with an independently justified estimate of current economic worth.
Official syllabus reference: [1]
20. Confirmation bias and asymmetric evidence standards
Confirmation bias involves seeking, interpreting, or remembering evidence in ways that favor an existing belief. The error can occur even when an analyst reads contrary information but demands much stronger proof from it. Adding several indicators derived from the same prices may reinforce conviction without providing independent confirmation. A substantive countermeasure is to specify what evidence would weaken the thesis and evaluate both supporting and opposing observations under comparable criteria. This differs from abandoning a view after every adverse observation; the issue is consistent treatment of evidence.
Worked example: Before issuing a bullish report, an analyst states that failure to hold a specified breakout would weaken the thesis. When that failure occurs, she evaluates it using the original rule rather than relabeling it harmless noise solely because the outlook was bullish.
Common mistake: Counting multiple agreeing indicators as independent evidence can conceal confirmation bias when all indicators largely transform the same underlying observations.
Official syllabus reference: [1]
21. Availability, recency, and base-rate neglect
Availability bias arises when ease of recall influences judgments about likelihood or importance. Recent, vivid, or widely reported events are often easier to retrieve than representative historical evidence. Recency can intensify this effect, but an event need not be recent to be memorable. Base-rate neglect occurs when general frequencies receive too little weight relative to a striking example. Apply the distinction by separating how memorable an event feels from how often comparable events occur, while recognizing that structural changes can make historical frequencies imperfect guides.
Worked example: After watching extensive coverage of one trading-system failure, an investor assumes all automated strategies are unusually dangerous. A more representative assessment compares failure rates, leverage, controls, and strategy types rather than allowing the most memorable case to determine the probability estimate.
Common mistake: Treating the easiest example to remember as the most probable outcome substitutes memory accessibility for representative evidence and relevant base rates.
Official syllabus reference: [1]
22. Overconfidence and self-attribution
Overconfidence can appear as exaggerated skill estimates, overly narrow forecast ranges, or excessive certainty about conclusions. Self-attribution can sustain it when investors credit successful outcomes to ability but explain failures through bad luck or outside interference. A profitable result alone cannot identify skill because market exposure and chance also contribute. Distinguish outcome quality from decision quality by examining the information available when the decision was made and performance across a meaningful set of observations. Calibrated confidence should reflect uncertainty rather than the emotional force of recent success.
Worked example: After three profitable trades during a broad rally, a trader doubles position sizes and claims exceptional forecasting ability. Comparing those trades with a passive benchmark and reviewing the original probabilities would help separate market exposure from evidence of repeatable skill.
Common mistake: Inferring superior skill from a short winning streak ignores chance, common market exposure, and selective explanations of unsuccessful trades.
Official syllabus reference: [1]
23. Herding and informational cascades
Herding occurs when participants align decisions with others' actions rather than relying fully on their own information. It can arise from social pressure, career concerns, or a belief that others possess superior information. An informational cascade develops when observed earlier choices outweigh later participants' private signals. Similar trades do not prove irrational herding because investors may independently respond to the same evidence. The analytical task is distinguishing shared information from dependence on others' behavior and recognizing that crowding can make positions vulnerable to a common exit.
Worked example: Several managers buy a fashionable sector because peer managers already own it, despite their own weak research findings. Their decisions reflect reputational and social incentives, while concentrated ownership may make a later change in sentiment more consequential for prices.
Common mistake: Labeling all correlated investor behavior herding ignores the possibility that independent analysis produces similar responses to the same public information.
Official syllabus reference: [1]
Classical Techniques
24. Arithmetic and logarithmic price scales
An arithmetic scale assigns equal vertical distance to equal absolute price changes. A logarithmic scale assigns equal distance to equal proportional changes because differences in logarithms represent ratios. A rise from $20 to $40 therefore matches the log-scale distance from $40 to $80, although the dollar changes differ. Scale selection can materially alter the appearance of long-term trendlines and acceleration. Logarithmic charts require positive values under the usual construction. State the scale when comparing long histories or interpreting whether a trend's rate of change appears stable.
Worked example: A stock doubles from $15 to $30 and later from $30 to $60. The moves occupy equal vertical distances on a logarithmic chart, helping an analyst compare percentage growth rather than mistaking the larger dollar move for faster proportional appreciation.
Common mistake: Reading a steeper arithmetic chart as proof of accelerating percentage growth ignores the effect of a rising price base.
Official syllabus reference: [1]
25. OHLC aggregation across chart intervals
An OHLC bar records the first traded price, highest price, lowest price, and final traded price within a defined interval. Aggregating daily bars into a weekly bar uses the first session's open, the maximum daily high, the minimum daily low, and the final session's close. Weekly volume is normally summed. Averaging daily opens or closes does not construct a valid weekly OHLC bar. Session definitions, holidays, and time zones matter because they determine which transactions belong in each interval and can alter apparent chart patterns.
Worked example: Across a three-session holiday week, the first open is $40, the highest daily high is $44, the lowest daily low is $39, and the last close is $43. Those four values form the weekly bar without averaging daily prices.
Common mistake: Building weekly bars from average daily prices changes the meaning of open, high, low, and close and can distort pattern interpretation.
Official syllabus reference: [1]
26. Candlestick bodies, shadows, and engulfing patterns
A candlestick body spans the open and close, while shadows extend to the period's high and low. A bullish engulfing pattern conventionally has a bullish real body that encompasses the preceding bearish real body; the shadows need not be engulfed. Its interpretation depends on preceding price behavior and location, especially whether it follows a decline. Candle color conventions can differ between charting platforms, so open and close values are decisive. The pattern suggests a change in buying and selling pressure without guaranteeing a sustained reversal or defining a complete trade.
Worked example: After a decline, one session opens at $32 and closes at $30. The next opens at $29.50 and closes at $32.50. The second real body engulfs the first, even if one earlier shadow extends beyond the newer candle's range.
Common mistake: Requiring both shadows to be engulfed confuses real-body engulfment with full-range containment and can lead to incorrect pattern identification.
Official syllabus reference: [1]
27. Point-and-figure box size and reversal settings
Point-and-figure charts emphasize price movement by recording rising columns of Xs and falling columns of Os according to specified box and reversal rules. Time does not advance at a uniform horizontal rate; a quiet period may add nothing. Box size controls the minimum recorded increment, while the reversal amount controls the price movement needed to start an opposing column. Larger settings filter more small fluctuations but delay recognition of some changes. Interpretation requires knowing whether boxes use fixed dollar, percentage, or other scaling and which price observations drive updates.
Worked example: With $1 boxes and a three-box reversal, an X column ending at $54 requires a decline sufficient to reach $51 before an opposing O column begins under the stated convention. A move only to $52 does not yet meet that reversal requirement.
Common mistake: Assuming each point-and-figure column represents one day ignores that columns are created by qualifying price movement rather than equal time intervals.
Official syllabus reference: [1]
28. Trendlines and parallel price channels
An uptrend line connects rising reaction lows; a downtrend line connects declining rally highs. A parallel line through opposite extremes can form a channel that describes the trend's approximate slope and boundaries. Additional well-spaced contacts can make a line more meaningful, but repeatedly redrawing it to fit every observation weakens its interpretive discipline. A break indicates that the prior slope is no longer being maintained; it does not alone establish a full reversal. Chart scale, selected pivots, and horizon affect the line's geometry and must remain consistent.
Worked example: A stock touches a rising support line at $40, $44, and $48 before closing below it at $47. The analyst records a break of the prior slope, then examines significant lows and subsequent rallies before declaring a primary downtrend.
Common mistake: Treating any trendline break as a complete reversal ignores that markets can slow, consolidate, or establish a shallower trend.
Official syllabus reference: [1]
29. Support, resistance, and polarity reversal
Support is a price area where buying has previously been sufficient to interrupt a decline; resistance is an area where selling has interrupted an advance. When price decisively crosses an area, its role may reverse: former resistance can become support and former support can become resistance. This polarity reflects changed positioning, expectations, and reference points, not a physical barrier. Treat levels as zones whose relevance depends on prior interaction and current behavior. A brief penetration, a sustained breakout, and a successful retest provide different amounts of evidence.
Worked example: A stock repeatedly stalls near $60, then closes above that area and later retreats to $60 before rebounding. The former resistance is now acting as support, although the observed retest does not guarantee that future declines will stop there.
Common mistake: Assuming support is an exact price that cannot be crossed overlooks market noise, gaps, liquidity, and the zone-based nature of historical trading interest.
Official syllabus reference: [1]
30. Breakaway, continuation, and exhaustion gaps
A price gap occurs when an interval's trading range does not overlap the preceding range under the chart convention used. Breakaway gaps leave an established congestion area; continuation gaps appear within an ongoing directional move; exhaustion gaps occur near a move's end. Their classification depends on location and follow-through, so it is often clearer afterward than at formation. Volume can add context but does not uniquely identify the gap type. Distinguish genuine trading gaps from artificial discontinuities caused by unadjusted corporate actions or incompatible data series.
Worked example: A stock opens above a six-week resistance area and trades entirely above the preceding day's high. Continued strength supports a breakaway interpretation; an immediate return into the range would instead weaken the breakout and require reassessment.
Common mistake: Assuming every gap must soon close confuses a common observation in some settings with a universal rule across gap types.
Official syllabus reference: [1]
31. Simple and exponential moving-average weighting
A simple moving average assigns equal weight to observations within its lookback and drops the oldest when a new observation arrives. An exponential moving average recursively weights the newest value and the previous average, commonly using α = 2/(n + 1). The EMA's older observations decay rather than disappearing abruptly. Both methods lag because they summarize past data, and responsiveness depends on settings and the price path. A crossover identifies a relationship between smoothed series, not an independent guarantee that a durable trend has begun.
Worked example: With a previous EMA of $100, a current close of $106, and α of one-third, the updated EMA is $102. The calculation moves the old average one-third of the way toward the latest close without eliminating the influence of earlier prices.
Common mistake: Calling an EMA lag-free because it emphasizes recent prices overlooks that recursive smoothing still incorporates historical observations and responds after price changes.
Official syllabus reference: [1]
32. Reversal-pattern completion and measured objectives
A head-and-shoulders top consists of a higher central peak between two lower peaks, with intervening lows defining a neckline. A double top has two comparable peaks separated by a reaction low. These are potential reversal structures until the relevant support is broken under the chosen confirmation rule. Conventional arithmetic objectives project pattern height from the breakout boundary, but sloping necklines require a stated measurement convention. Pattern identification, completion, and target estimation are separate steps. A target is conditional and does not guarantee either execution or eventual price arrival.
Worked example: A double top forms near $80 with an intervening low at $72. After a qualifying break below $72, its $8 height gives an indicative $64 objective when projected downward from support using the conventional arithmetic method.
Common mistake: Calling two similar peaks a completed double top before support breaks confuses a possible formation with a confirmed pattern under the stated method.
Official syllabus reference: [1]
33. Triangles: converging boundaries and breakout direction
Triangles contain price swings between converging boundaries. An ascending triangle combines relatively flat resistance with rising lows; a descending triangle combines relatively flat support with falling highs; a symmetrical triangle has both boundaries converging. These structures describe compression and competing pressure, but their names do not guarantee breakout direction. The actual boundary break and subsequent behavior determine whether the anticipated continuation or reversal develops. A conventional objective may use the triangle's widest height, with the projection anchor and confirmation rule stated explicitly.
Worked example: A stock repeatedly reaches $50 while reaction lows rise from $43 to $45 to $47. The structure resembles an ascending triangle, but a break below the rising lower boundary would contradict an assumed bullish resolution rather than confirm one.
Common mistake: Buying solely because a triangle is called ascending substitutes a pattern label for evidence that resistance has actually been overcome.
Official syllabus reference: [1]
34. Flags and pennants as consolidation structures
Flags and pennants are brief consolidations that commonly follow a sharp directional move called the flagpole. A flag often has roughly parallel boundaries, while a pennant contracts between converging boundaries. They are usually interpreted as possible continuation structures, but continuation remains conditional on a breakout and follow-through. A small triangle without a preceding impulsive move lacks an important part of the pennant context. Measured objectives sometimes project the flagpole length from the breakout, yet this is an estimate rather than a promise about magnitude or timing.
Worked example: A stock advances rapidly from $40 to $50, then trades in a narrow downward-sloping channel. A later break above that channel supports a bullish flag interpretation more strongly than a similar channel occurring without the preceding sharp advance.
Common mistake: Labeling every small consolidation a flag or pennant ignores the preceding directional impulse that gives the continuation interpretation its context.
Official syllabus reference: [1]
35. RSI: gain–loss balance and divergence
The Relative Strength Index transforms the ratio of smoothed average gains to smoothed average losses into a bounded oscillator: RSI = 100 − 100/(1 + RS). It measures an instrument's internal gain–loss balance, not performance against a benchmark. A bearish divergence occurs when price makes a higher high while RSI makes a lower high, suggesting reduced momentum. Elevated readings can persist during strong trends, so a commonly used overbought boundary is not an automatic sell instruction. Lookback and smoothing conventions affect values and should be specified.
Worked example: A stock rises from a prior peak of $70 to a new peak of $74, while RSI falls from 76 to 68 at the corresponding highs. The analyst identifies bearish momentum divergence without claiming that the price uptrend has already reversed.
Common mistake: Confusing RSI with a stock-to-index relative-strength ratio mixes an internal momentum oscillator with a measure of comparative asset performance.
Official syllabus reference: [1]
36. MACD line, signal line, and histogram
MACD is commonly constructed as a shorter exponential moving average minus a longer exponential moving average. A signal line smooths MACD, and the histogram is MACD minus that signal line. A positive MACD means the shorter average exceeds the longer; a rising histogram means the difference between MACD and its signal line is increasing. These are distinct comparisons. A histogram can rise while remaining negative, indicating improving momentum relative to the signal line rather than necessarily positive price returns. Parameters vary and should not be treated as universally optimal.
Worked example: MACD moves from −1.2 to −0.8 while its signal line moves from −0.7 to −0.75. The histogram improves from −0.5 to −0.05, but MACD remains below zero and below the signal line at the second observation.
Common mistake: Interpreting any rising histogram as a completed bullish crossover ignores whether the histogram has actually crossed above zero.
Official syllabus reference: [1]
37. Stochastic oscillator and closing-range location
The stochastic oscillator locates the current close within a recent high–low range. A basic %K calculation is 100 × (close − lowest low)/(highest high − lowest low), with smoothing variants producing additional lines. A high reading means the close is near the top of that range, not that the asset has outperformed another asset. A flat range with equal high and low requires a defined handling convention because the denominator is zero. Extreme readings can persist in trends, so crossing and divergence signals require context.
Worked example: Over the selected lookback, the highest high is $60 and the lowest low is $50. A $58 close gives %K = 80, placing the close four-fifths of the way up the range rather than establishing an 80% probability of a decline.
Common mistake: Reading a stochastic value of 80 as an 80% reversal probability mistakes a normalized range position for a calibrated statistical forecast.
Official syllabus reference: [1]
38. On-balance volume and cumulative signed activity
On-balance volume adds an interval's volume when its close exceeds the prior close, subtracts volume when the close is lower, and commonly leaves the total unchanged when closes are equal. It assigns the entire interval's volume according to close-to-close direction rather than identifying each trade's aggressor. The absolute starting value is arbitrary, so the line's trend and divergences matter more than its level. OBV can provide participation context, but it does not measure actual net ownership changes or prove that institutions are accumulating a position.
Worked example: Starting from an arbitrary OBV of zero, an up-close day with 1.2 million shares adds 1.2 million. A following down-close day with 0.8 million subtracts that amount, leaving OBV at 0.4 million despite both sessions having buyers and sellers.
Common mistake: Treating OBV as measured institutional money flow overstates what can be inferred from signing total volume by the direction of the close.
Official syllabus reference: [1]
39. Advance–decline breadth and index concentration
An advance–decline line cumulatively adds advancing issues minus declining issues for a defined universe. It measures participation differently from a capitalization-weighted price index, which can be dominated by large constituents. A new index high without a corresponding breadth high suggests narrower participation, while improving breadth can show broader support. Divergence is a warning rather than conclusive reversal evidence. Universe selection, unchanged issues, listing changes, and the inclusion of non-operating securities can affect interpretation, so comparisons should use consistent breadth definitions and appropriate index coverage.
Worked example: On one session, 700 issues advance and 500 decline, adding 200 to the cumulative line. If several large constituents lift the index while later sessions show negative breadth, the index can strengthen even as the participation measure deteriorates.
Common mistake: Assuming a rising capitalization-weighted index means most constituents are advancing overlooks how concentrated weights can separate index direction from breadth.
Official syllabus reference: [1]
40. TRIN and average volume per advancing issue
TRIN divides the advancing-to-declining issue ratio by the advancing-to-declining volume ratio. Algebraically, it also equals average volume per declining issue divided by average volume per advancing issue. A reading above one therefore indicates greater volume per declining issue under that calculation, not necessarily more declining issues overall. Zero denominators and inconsistent exchange coverage require explicit treatment. Intraday readings can be noisy, and historical smoothing or contextual interpretation is preferable to assuming a universal threshold that automatically identifies a market turning point.
Worked example: With 400 advancing issues carrying 200 million shares and 200 declining issues carrying 150 million shares, TRIN is 2 divided by 1.333, or 1.5. Average volume is 500,000 shares per advancer versus 750,000 per decliner.
Common mistake: Interpreting TRIN above one as proof that decliners outnumber advancers ignores the separate volume ratio built into the indicator.
Official syllabus reference: [1]
41. Put–call ratios and external sentiment indicators
A put–call ratio compares put activity with call activity, using a stated basis such as contract volume or open interest. It provides information outside the analyzed security's own price series and can help assess sentiment or hedging demand. Higher put activity may reflect bearish speculation, protective hedging, or institutional portfolio practices; it is not a direct count of bearish investors. Equity-only, index-only, and combined ratios can behave differently. Compare consistent series and historical distributions before interpreting an extreme as a potential contrary signal.
Worked example: An index put–call volume ratio rises during a portfolio-hedging period while an equity-only ratio changes little. The analyst distinguishes the two participant contexts instead of treating the index reading as direct evidence that individual stock traders have become uniformly bearish.
Common mistake: Assuming every put purchase expresses a standalone bearish forecast ignores hedging, spreads, and the different motives represented in options activity.
Official syllabus reference: [1]
42. Cycle period, frequency, amplitude, and phase
Period is the time required for one complete cycle, while frequency is its reciprocal in cycles per unit time. Amplitude measures displacement from a midpoint, and phase specifies position relative to a reference cycle. A timing shift changes phase without necessarily changing period or amplitude. These definitions describe idealized oscillations; actual market cycles can drift, weaken, or overlap. State units and the reference point when measuring them. A price series with recurring peaks is not automatically a stable cycle capable of producing precise future turning dates.
Worked example: An idealized cycle repeats every 20 sessions, giving frequency 0.05 cycle per session. A second identical wave shifted five sessions later has the same amplitude and period but a quarter-cycle phase lag relative to the first.
Common mistake: Using amplitude to describe a timing offset confuses vertical magnitude with phase, while using frequency for elapsed duration reverses the reciprocal relationship.
Official syllabus reference: [1]
43. Cycle superposition and apparent irregularity
Superposition means that observed movement can be represented as the sum of multiple components. Cycles of different periods and phases can reinforce one another or partially cancel, creating an irregular-looking combined series even when each component is simple. A longer trend can further obscure the oscillations. Decomposition can help describe historical structure, but fitting many components risks explaining noise and does not establish stable forecasting power. Use the principle to understand why observed peaks may vary in height and spacing rather than expecting a single perfect market rhythm.
Worked example: A 20-session oscillation peaks while a 40-session oscillation is near its trough, producing a muted combined high. Later their peaks align and the combined movement is larger, even though neither component's individual amplitude has changed.
Common mistake: Interpreting every unequal peak as a changed underlying cycle overlooks that overlapping components can alter the combined series through reinforcement and cancellation.
Official syllabus reference: [1]
Advanced Techniques
44. Simple returns, log returns, and compounding
A simple return is ending value divided by beginning value minus one. Multi-period wealth compounds through products of one plus each simple return, so arithmetic addition generally fails. A log return is the natural logarithm of the price ratio and adds across periods for a positive-price series. Log and simple returns are close for small changes but diverge for large moves. Price returns exclude distributions unless the series includes them, so state whether the analysis uses price or total return and keep adjustments consistent.
Worked example: A $100 investment gains 20% to $120 and then loses 20% to $96. Its compounded return is −4%, not zero. The two-period wealth relative is 1.20 × 0.80 = 0.96, illustrating why equal percentage gains and losses do not cancel.
Common mistake: Adding ordinary percentage returns across periods ignores the changing investment base and can materially misstate compound growth after volatile movements.
Official syllabus reference: [1]
45. Mean, median, and outlier sensitivity
The arithmetic mean sums observations and divides by their number, making it sensitive to extreme values. The median identifies the middle ranked observation, or the average of the two middle values for an even sample, and is more resistant to outliers. Neither statistic is universally superior: mean return relates to expected single-period payoff, while the median describes a typical ranked observation. A difference between them can reveal skewness but does not fully describe the distribution. Inspect dispersion and tail observations rather than reducing performance to one central statistic.
Worked example: Five trade returns are −2%, −1%, 0%, 1%, and 17%. Their median is 0%, while their arithmetic mean is 3%. The strategy's positive average depends heavily on one large winner, a feature the median alone would conceal.
Common mistake: Discarding an outlier merely because it changes the mean can remove a genuine feature of a strategy's payoff distribution.
Official syllabus reference: [1]
46. Variance, standard deviation, and sample denominators
Variance averages squared deviations from a mean, while standard deviation takes the square root and returns the measure to the original units. A population variance uses the population size as denominator; a conventional sample variance estimating population dispersion uses n − 1. State which interpretation applies before calculating. Standard deviation measures overall dispersion and treats upside and downside deviations symmetrically. It is not maximum loss, a guaranteed trading range, or a complete tail-risk measure. Estimated volatility can also change materially with lookback and return frequency.
Worked example: For sample returns of −1%, 1%, and 3%, the mean is 1%. Squared deviations sum to 8 percentage-points squared. Dividing by two gives sample variance 4 and sample standard deviation 2 percentage points for the observations.
Common mistake: Interpreting standard deviation as the largest possible loss confuses a dispersion estimate with a bound on extreme outcomes.
Official syllabus reference: [1]
47. Normal approximations, skewness, and fat tails
A normal model summarizes a symmetric distribution through its mean and standard deviation, but market returns can exhibit asymmetry, heavy tails, and changing volatility. Skewness concerns asymmetry; fat tails indicate more extreme observations than a comparable normal model would suggest. A standardized score measures distance from the mean in standard-deviation units, but converting it into a probability requires distributional assumptions. Historical frequencies also carry sampling uncertainty. Apply normal approximations as models with stated limitations rather than treating rare modeled events as impossible market outcomes.
Worked example: A daily return lies four estimated standard deviations below the historical mean. The analyst reports its standardized magnitude but avoids assigning a normal-model probability as an established real-world frequency without checking tail behavior and volatility stability.
Common mistake: Calling an extreme return impossible because a normal model assigns it low probability mistakes a convenient approximation for a guaranteed description of markets.
Official syllabus reference: [1]
48. Correlation, regression slope, and causation
Correlation measures the direction and strength of a linear relationship on a standardized scale, while a regression slope measures the expected change in one variable associated with a unit change in another under the fitted model. A slope depends on units; correlation does not change when positive unit conversions are applied. Neither statistic establishes causation. Relationships estimated from nonstationary price levels can be misleading, so returns or another justified transformation may be more appropriate. Outliers, changing regimes, and nonlinear dependence can also limit interpretation of one summary coefficient.
Worked example: Two assets have strongly correlated daily returns, but one usually moves twice as much as the other. Their correlation describes common direction, while a fitted slope captures relative sensitivity; the relationship alone does not show that trading one causes the other to move.
Common mistake: Treating high correlation as equal risk or proof of causation overlooks differences in volatility, units, omitted drivers, and the possibility of changing dependence.
Official syllabus reference: [1]
49. True range and Wilder-style ATR smoothing
True range is the maximum of the current high–low distance and the absolute distances from the previous close to the current high and low. Including the previous close captures gaps missed by the intraday range. Average true range smooths these observations; a common Wilder update is ATR(new) = ((n − 1) × ATR(previous) + TR)/n after initialization. ATR measures movement magnitude in price units without indicating direction. Dollar ATR is not directly comparable across differently priced assets unless normalized or otherwise adjusted for the analytical purpose.
Worked example: With a prior 14-period ATR of $2 and a new true range of $3.40, the Wilder-style update is (13 × $2 + $3.40)/14 = $2.10. The higher reading indicates greater recent range, not a bullish forecast.
Common mistake: Reading rising ATR as upward momentum confuses a directionless measure of price movement with a directional trend or return indicator.
Official syllabus reference: [1]
50. Bollinger Bands and volatility compression
Bollinger Bands commonly place upper and lower bands a specified number of rolling price standard deviations around a moving average. Their width expands and contracts with the chosen dispersion estimate. Narrowing bands indicate lower recent measured dispersion, but they do not determine the next breakout's direction. A band touch is not automatically a reversal because prices can travel near an outer band during a trend. The familiar normal-distribution coverage interpretation does not automatically apply to dependent, changing market prices around a rolling estimated center.
Worked example: A moving average is $100 and the specified rolling standard deviation falls from $3 to $1. With bands two standard deviations away, the interval narrows from $94–$106 to $98–$102, showing compression without selecting an upward or downward resolution.
Common mistake: Assuming a squeeze predicts a rally confuses declining measured volatility with directional evidence that the band calculation does not supply.
Official syllabus reference: [1]
51. Realized versus implied volatility and time scaling
Realized volatility summarizes observed historical variability, while implied volatility is inferred from option prices through a pricing framework. They differ in information, measurement, and interpretation; implied volatility is not a guaranteed forecast of subsequent realized movement. Under an explicit independent, constant-variance approximation, volatility scales with the square root of time. The time convention must match the quoted measure. Scaling does not describe the maximum possible move or the expected direction, and jumps, dependence, or changing variance can make the approximation inaccurate.
Worked example: For an exercise assuming 252 trading days and constant independent daily variance, 1% daily volatility annualizes to 1% × √252, approximately 15.87%. The calculation changes the horizon of the dispersion estimate without predicting a 15.87% annual gain.
Common mistake: Treating annualized volatility as an expected return or maximum loss assigns directional and bounding meanings that a dispersion estimate does not have.
Official syllabus reference: [1]
52. Systematic rules: signals, filters, and execution
A systematic strategy expresses decisions through reproducible rules. A signal identifies a potential action, a filter determines whether conditions permit it, and execution rules specify when and how an order can occur. Position sizing and exit logic complete the operational description. For example, a breakout trigger can be restricted to an upward market regime without changing the distinction between trigger and filter. Rules must use information available at decision time and account for practical prices and costs. Deterministic rules improve reproducibility but do not guarantee profitability or eliminate model risk.
Worked example: A strategy buys after a 20-session closing high only when the benchmark exceeds its 100-session average, then executes at the next session's modeled open. The breakout is the trigger, the benchmark condition is the filter, and the next-open assumption defines timing.
Common mistake: Calling an indicator a complete trading system omits the decision conditions, executable timing, position size, and exits needed to reproduce outcomes.
Official syllabus reference: [1]
53. Look-ahead and survivorship bias in historical tests
Look-ahead bias occurs when a test uses information before it would have been available, including revised data or an execution price formed before a required signal was known. Survivorship bias arises when the historical universe excludes securities that later disappeared or failed. Both can exaggerate apparent performance for different reasons: impossible foresight and selective membership. Use point-in-time data, historical constituents, appropriate corporate-action treatment, and realistic decision timestamps. A favorable test remains unpersuasive if its information sequence or security universe could not have existed in real time.
Worked example: A researcher tests today's successful index constituents back ten years using their later membership and same-close signals. Rebuilding historical membership addresses survivorship, while moving execution after signal availability addresses timing; repairing one defect does not automatically repair the other.
Common mistake: Assuming adjusted prices alone make a backtest unbiased overlooks historical membership, publication lags, revised information, and whether modeled trades were executable.
Official syllabus reference: [1]
54. Out-of-sample testing and parameter robustness
An out-of-sample segment is withheld from strategy construction and parameter selection to assess performance on previously unused observations. Repeatedly inspecting and retuning against it turns it into part of the development process. Robustness concerns whether results remain reasonably consistent across nearby parameters, plausible costs, and relevant environments rather than depending on one isolated optimum. Time-ordered evaluation helps preserve chronology, but no validation method guarantees future success. Record how many alternatives were explored because extensive searching increases the probability of selecting a result that fits historical noise.
Worked example: A moving-average strategy performs well only at a 37-session lookback, while 35, 36, 38, and 39 perform poorly. That isolated optimum is less reassuring than a broad region of acceptable results followed by success on untouched later data.
Common mistake: Retuning after every holdout disappointment while continuing to call the same data out of sample conceals repeated information leakage into model selection.
Official syllabus reference: [1]
55. Relative-strength ratios and comparative returns
A relative-strength ratio divides one asset or index by another using comparable observations. The ratio rises when the numerator outperforms the denominator, even if both decline. Its period return is (1 + numerator return)/(1 + denominator return) − 1, which differs from simple return subtraction except as an approximation. Price ratios and total-return ratios answer different questions when distributions matter. Currency, timing, and index construction must also be aligned. This comparative measure is distinct from RSI and does not establish whether either asset offers a positive absolute return.
Worked example: A sector loses 5% while its benchmark loses 10%. The ratio changes by 0.95/0.90 − 1, approximately +5.56%. The sector is relatively strong but still produces a negative absolute return for an unhedged holder.
Common mistake: Interpreting a rising relative-strength line as proof of an absolute gain ignores that outperformance can result from losing less than the benchmark.
Official syllabus reference: [1]
56. Intermarket relationships and macroeconomic regimes
Intermarket analysis examines relationships among equities, bonds, commodities, currencies, and their underlying economic drivers. Bond prices and yields generally move inversely for a fixed set of promised cash flows, while relationships across different asset classes depend on inflation, growth, policy, and risk preferences. Equities and bonds may fall together during an inflation shock yet move oppositely during a growth scare. Use cross-asset observations to test an economic interpretation rather than assuming permanent correlations or fixed lead–lag rules. Comparable horizons and currency effects are essential to meaningful comparisons.
Worked example: Bond prices decline while commodity prices rise and equities weaken. An analyst considers an inflation-driven explanation, then compares it with other evidence. She does not apply a rule that falling bonds must always precede the same equity-market outcome.
Common mistake: Applying a historical stock–bond relationship as a permanent law ignores changing macroeconomic drivers and the possibility that correlations reverse across regimes.
Official syllabus reference: [1]
57. Trade expectancy, payoff ratio, and profit factor
Expected trade profit combines outcome probabilities and average payoffs: win probability times average win minus loss probability times average loss, with consistent cost treatment. A high win rate can coexist with negative expectancy when occasional losses are large. Profit factor divides aggregate gross profits by the absolute aggregate gross losses under the stated convention; it is not the percentage of winning trades. Historical estimates carry sampling uncertainty and may conceal clustering or tail exposure. Compare strategies using the full payoff structure rather than selecting whichever wins most frequently.
Worked example: A strategy wins 40% of trades, averaging $300 per win, and loses 60%, averaging $150 per loss. Before additional costs, expectancy is $120 − $90 = $30 per trade, demonstrating that a minority win rate can still support positive expectancy.
Common mistake: Choosing the highest win-rate strategy without examining loss size and costs ignores the payoff asymmetry that determines expected profitability.
Official syllabus reference: [1]
58. Position risk, drawdown, and recovery arithmetic
Risk-based position sizing divides a stated monetary risk budget by planned loss per unit, including appropriate cost and execution allowances. It controls a modeled exposure rather than guaranteeing realized loss, because gaps and liquidity can defeat stop assumptions. Drawdown measures decline from a prior equity peak; recovery requires a larger percentage gain than the preceding percentage loss. After fractional drawdown d, the gain needed to recover is d/(1 − d). Portfolio risk also depends on correlated positions, so several individually sized trades can collectively exceed an intended exposure.
Worked example: An account falls from $100,000 to $80,000, a 20% drawdown. Recovering $20,000 from the smaller base requires 25%, calculated as 0.20/0.80. This asymmetry explains why controlling large losses matters even when future percentage gains appear comparable.
Common mistake: Assuming equal percentage gains and losses restore equity ignores the smaller capital base after a drawdown and the larger recovery percentage required.
Official syllabus reference: [1]
Ethics
59. Independence, objectivity, and conflict disclosure
Professional ethics requires analytical judgments that are not improperly influenced by personal interests, compensation, gifts, or relationships. A relevant conflict should be avoided where appropriate or disclosed clearly enough for affected readers and clients to understand it. Internal permission for an activity does not automatically remove its disclosure implications. Conversely, disclosure does not authorize biased analysis or conduct otherwise prohibited. Apply these principles to research ownership, issuer relationships, and incentives by distinguishing the existence of an interest from its effect on judgment and the transparency required for informed evaluation.
Worked example: An analyst owns shares covered by a technical report and receives no issuer payment. She discloses the relevant ownership prominently and bases the recommendation on documented analysis, recognizing that permitted personal ownership can still affect how readers assess potential conflicts.
Common mistake: Treating a generic investment-risk disclaimer as conflict disclosure hides the specific personal interest readers need to evaluate the analyst's independence.
Official syllabus reference: [1]
60. Competence, reasonable basis, and honest representation
Ethical technical research requires sufficient competence to perform the work and a reasonable basis for recommendations. Analysts should understand material data limitations, calculation methods, and assumptions before presenting results as reliable. Honest representation distinguishes observed facts, analytical opinions, hypothetical tests, and actual performance; it also acknowledges others' work rather than implying authorship. Competence is an ongoing professional responsibility, not something established permanently by an examination result. Apply these principles when using unfamiliar software or models by verifying substantive outputs and communicating limitations that materially affect interpretation.
Worked example: A vendor tool reports strong historical strategy returns, but the analyst discovers that delisted securities were excluded. She identifies the survivorship limitation and does not present the simulation as actual client performance or as validated evidence of a dependable future return.
Common mistake: Assuming a polished vendor output supplies a reasonable research basis overlooks the analyst's responsibility to understand material assumptions and represent results accurately.
Official syllabus reference: [1]