Time Series Analysis Cheat Sheet 2026

The 30 highest-yield Time Series Analysis facts, distilled from real exam questions. Print it, save it as a PDF, or study it here — free, no sign-up.

1. In classical decomposition, what is the 'centered moving average' used to estimate? → The trend-cycle component
2. What is ARCH (Autoregressive Conditional Heteroscedasticity) in time series modeling? → A model where the variance of the error term depends on past squared errors
3. What is the difference between the 'trend' and 'cycle' components in time series decomposition? → Trend is a long-term direction; cycle is a medium-term fluctuation around the trend
4. Which metric is commonly minimized to select between competing exponential smoothing models? → AICc (corrected Akaike Information Criterion)
5. What is a 'long memory' process in time series analysis? → A process where autocorrelations decay slowly (hyperbolically) rather than exponentially
6. What does a negative autocorrelation at lag 1 in a time series indicate? → High values tend to be followed by low values and vice versa
7. What does a 'structural break' mean for stationarity testing? → A sudden shift in level or trend causes standard unit root tests to have low power
8. We can't find trend values of some things using the moving average method. → Starting and End Periods
9. How are the smoothing parameters (α, β, γ) typically estimated in exponential smoothing models? → By minimizing the sum of squared one-step-ahead forecast errors
10. What does a 'remainder' (residual) component after decomposition ideally look like? → White noise with no systematic patterns
11. In time series forecasting, what is the 'horizon' h? → The number of steps ahead being forecast
12. If the ACF of a time series shows a sinusoidal pattern with significant spikes at regular intervals, what does this most likely indicate? → The series contains a seasonal component
13. What is the 'naive' forecasting method? → Using the most recent observation as the forecast for all future periods
14. What is the main advantage of STL over classical additive decomposition? → STL handles any type of seasonality and is robust to outliers
15. What does the PACF plot help identify in an ARIMA model? → The AR order 'p'
16. What does the seasonal index represent in classical decomposition? → The average deviation from the trend for each period within a season
17. What is a 'prediction interval' in forecasting? → A range within which a future observation will fall with a specified probability
18. For which type of series is simple exponential smoothing most appropriate? → Non-seasonal data with no systematic trend
19. How does seasonal differencing differ from regular differencing? → Seasonal differencing subtracts the value from the same period in a prior season
20. What is 'dynamic time warping' (DTW) used for in time series analysis? → Measuring similarity between two time series that may be shifted or distorted in time
21. Which visual tool is most helpful for an initial stationarity assessment? → Time series plot showing mean and variance over rolling windows
22. In the Box-Jenkins methodology, which tools are primarily used in the model identification stage? → ACF and PACF plots
23. The secular tendency has undergone the following movement(s). → All of the above
24. What is the purpose of an Intervention Analysis in time series? → To model the effect of known external events (interventions) on a time series
25. For a pure AR(p) process, which of the following correctly describes the PACF? → It cuts off to zero after lag p
26. How many cycles does business get? → Four stages
27. What does the autocorrelation function (ACF) measure in a time series? → The correlation between a series and a lagged version of itself
28. The Ljung-Box test statistic is used to: → Test whether a group of autocorrelations are jointly zero
29. What does the 'I' in ARIMA stand for? → Integrated
30. The KPSS test differs from the ADF test in which fundamental way? → KPSS null hypothesis is stationarity; ADF null is non-stationarity