Certified Analytics Professional (CAP): key terms

Flashcards for this exam Certified Analytics Professional (CAP) guide

• Alternative Hypothesis (H1)The claim that there is an effect or difference, accepted when the null hypothesis is rejected.Foundational
• Analysis of Variance (ANOVA)A test of whether the means of three or more groups differ, by comparing between-group variance with within-group variance (the F-statistic).Intermediate
• ARIMAAutoRegressive Integrated Moving Average: a forecasting model that combines autoregression, differencing (to reach stationarity) and moving-average error terms.Advanced
• Bayes' TheoremA rule for updating the probability of a hypothesis given evidence: P(A|B) = P(B|A) × P(A) / P(B).Intermediate
• Binomial DistributionThe distribution of the number of successes in n independent trials, each with the same success probability p.Intermediate
• BootstrappingEstimating the sampling distribution of a statistic by repeatedly resampling the observed data with replacement.Advanced
• Box PlotA chart showing a distribution's median, quartiles (the box), whiskers and outliers; also called a box-and-whisker plot.Foundational
• Central Limit Theorem (CLT)The theorem that the distribution of sample means approaches a normal distribution as sample size grows, regardless of the population's shape.Intermediate
• Chi-Square Test (χ²)A test for categorical data: goodness of fit (observed vs expected frequencies) or independence (association between two categorical variables).Intermediate
• Cluster SamplingDividing the population into clusters (such as schools or stores), randomly selecting whole clusters, and sampling everyone (or a sample) within them.Intermediate
• Coefficient of Variation (CV)The standard deviation divided by the mean, often given as a percentage, used to compare variability across datasets with different scales.Intermediate
• Cohort AnalysisGrouping users by a shared starting event or period, such as signup month, and tracking their behavior over time.Intermediate
• Conditional ProbabilityThe probability of an event given that another event has occurred: P(A|B) = P(A and B) / P(B).Intermediate
• Confidence Interval (CI)A range computed from sample data that, under repeated sampling, would contain the true population parameter a stated percentage of the time.Intermediate
• Confounding VariableA variable that influences both the supposed cause and the effect, creating a spurious association.Intermediate
• Correlation CoefficientPearson's r: a value from -1 to +1 measuring the strength and direction of a linear relationship between two numeric variables.Intermediate
• Correlation vs CausationThe principle that a statistical association between two variables does not prove that one causes the other.Foundational
• CRISP-DMCross-Industry Standard Process for Data Mining: six iterative phases from business understanding through data prep, modeling, evaluation and deployment.Intermediate
• Descriptive AnalyticsAnalytics that answers 'what happened?' by summarizing historical data through reports, dashboards and KPIs.Foundational
• Descriptive StatisticsStatistics that summarize and describe the features of a dataset, such as its center, spread and shape, without generalizing beyond it.Foundational
• Diagnostic AnalyticsAnalytics that answers 'why did it happen?' through drill-down, correlation, segmentation and root-cause analysis.Foundational
• Effect SizeA measure of the magnitude of a difference or relationship, independent of sample size, such as Cohen's d or a correlation coefficient.Advanced
• Empirical RuleThe 68-95-99.7 rule: in a normal distribution about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations of the mean.Foundational
• Expected Value (EV)The probability-weighted average of all possible outcomes of a random variable: E[X] = Σ x × P(x).Intermediate
• Exploratory Data Analysis (EDA)An open-ended investigation of a dataset with summary statistics and visualizations to understand structure, spot anomalies and form hypotheses.Foundational
• Exponential SmoothingA forecasting method that weights observations with exponentially decreasing weights, with variants for trend (Holt) and seasonality (Holt-Winters).Advanced
• Five-Number SummaryA distribution summary consisting of the minimum, Q1, median, Q3 and maximum.Foundational
• HeteroscedasticityNon-constant variance of regression residuals across the range of predicted values.Advanced
• Hypothesis TestingA statistical procedure for deciding whether sample evidence is strong enough to reject a null hypothesis about a population.Intermediate
• Inferential StatisticsUsing sample data to draw conclusions or make predictions about a larger population, with quantified uncertainty.Foundational
• Interquartile Range (IQR)The spread of the middle 50% of data: the third quartile minus the first quartile (Q3 - Q1).Foundational
• KurtosisA measure of the tailedness of a distribution relative to a normal distribution.Advanced
• Levels of MeasurementThe four scales of data: nominal (labels), ordinal (ordered categories), interval (equal spacing, no true zero) and ratio (equal spacing with a true zero).Foundational
• Margin of Error (MOE)Half the width of a confidence interval: the critical value multiplied by the standard error.Foundational
• MeanThe arithmetic average: the sum of values divided by the number of values.Foundational
• Measures of Central TendencySingle values that describe the center of a distribution: mean, median and mode.Foundational
• MedianThe middle value of ordered data, or the average of the two middle values when the count is even.Foundational
• Misleading VisualizationA chart that distorts the data's message through truncated or inconsistent axes, inappropriate scales, cherry-picked ranges or area distortion.Foundational
• Moving AverageThe average of the most recent n periods, recalculated each period to smooth short-term fluctuations and reveal trends.Foundational
• MulticollinearityHigh correlation among predictor variables in a regression, which makes coefficient estimates unstable and hard to interpret.Advanced
• Multiple Linear RegressionA regression model that predicts a numeric outcome from two or more predictors: y = β0 + β1x1 + ... + βkxk + ε.Intermediate
• Normal DistributionA symmetric, bell-shaped continuous probability distribution fully described by its mean and standard deviation.Foundational
• Null Hypothesis (H0)The default assumption of no effect, no difference or no relationship, which a test tries to find evidence against.Foundational
• One-Tailed TestA hypothesis test in which the alternative specifies a direction (greater than or less than), so the rejection region lies in one tail.Intermediate
• Ordinal DataCategorical data with a meaningful order but unequal or unknown spacing between categories.Foundational
• OutlierAn observation that lies unusually far from the other values in a dataset.Foundational
• P-ValueThe probability, assuming the null hypothesis is true, of observing results at least as extreme as those in the sample.Intermediate
• Paired t-TestA t-test on the differences between paired measurements of the same subjects, such as before and after a treatment.Advanced
• PercentileThe value below which a given percentage of observations fall; the 90th percentile exceeds 90% of the values.Foundational
• Poisson DistributionThe distribution of the count of events in a fixed interval of time or space, when events occur independently at a constant average rate λ.Intermediate
• Population and SampleThe population is the entire group of interest; a sample is the subset actually observed and measured.Foundational
• Predictive AnalyticsAnalytics that answers 'what is likely to happen?' using statistical models and machine learning on historical data.Foundational
• Prescriptive AnalyticsAnalytics that answers 'what should we do?' by recommending actions using optimization, simulation and decision models.Intermediate
• Probability DistributionA function describing the likelihood of each possible value of a random variable.Foundational
• QuartileValues that split ordered data into four equal parts: Q1 (25th percentile), Q2 (the median) and Q3 (75th percentile).Foundational
• ResidualThe difference between an observed value and the value predicted by a model (y - ŷ).Intermediate
• Sampling BiasSystematic error that occurs when some members of a population are more likely to be sampled than others, so the sample misrepresents the population.Foundational
• Scatter PlotA chart that plots two numeric variables as points to reveal relationships, clusters and outliers.Foundational
• SeasonalityRegular, calendar-related patterns that repeat over fixed periods, such as daily, weekly or yearly cycles.Foundational
• Selection BiasBias introduced when the way subjects or data are selected relates to the outcome being studied.Intermediate
• Significance Level (α)The threshold probability α for rejecting the null hypothesis, commonly 0.05, which equals the accepted Type I error rate.Foundational
• Simple Random Sampling (SRS)A sampling method in which every member of the population has an equal chance of being selected.Foundational
• Simpson's ParadoxA phenomenon in which a trend in several groups reverses or disappears when the groups are combined.Advanced
• SkewnessA measure of a distribution's asymmetry: right (positive) skew has a long right tail, left (negative) skew a long left tail.Intermediate
• Spearman Rank Correlation (ρ)A nonparametric correlation computed on ranks, measuring the strength of a monotonic relationship.Advanced
• Standard Deviation (SD)The square root of variance, describing typical distance from the mean in the data's own units.Foundational
• Standard Error (SE)The standard deviation of a sampling distribution, usually of the mean: SE = s / √n.Intermediate
• Statistical PowerThe probability that a test correctly rejects a false null hypothesis (1 - β); 80% is a common target.Advanced
• Student's t-TestA test comparing means (one sample against a value, or two independent groups) when the population standard deviation is unknown, using the t-distribution.Intermediate
• Survivorship BiasDrawing conclusions only from entities that 'survived' a selection process while ignoring those that failed or dropped out.Intermediate
• Systematic SamplingSelecting every kth item from an ordered list after a random start.Intermediate
• Time SeriesA sequence of observations recorded at successive points in time, often at regular intervals.Intermediate
• Time Series DecompositionSeparating a time series into trend, seasonal and residual (irregular) components, additively or multiplicatively.Advanced
• Two-Tailed TestA hypothesis test in which the alternative is non-directional (≠), splitting α across both tails of the distribution.Intermediate
• Type I ErrorRejecting a true null hypothesis: a false positive, with probability α.Intermediate
• Type II ErrorFailing to reject a false null hypothesis: a false negative, with probability β.Intermediate
• VarianceThe average squared deviation from the mean, measuring how spread out data are; units are the square of the data's units.Intermediate
• Z-ScoreThe number of standard deviations a value lies from the mean: z = (x - μ) / σ.Intermediate
• Z-TestA hypothesis test using the standard normal distribution, appropriate when the population SD is known or samples are large, including tests of proportions.Intermediate