Module 2 - Intro to Time Series Certification Review Exam Questions And Answers Verified Solutions.

time series - correct answer a sequence of data points that is evenly spaced over time 4 time series components - correct answer trend, seasonal, cyclical, random trend component - correct answer - persistent overall upward or downward pattern - typically several years in duration - doesn't have to be linear - may change direction seasonal component - correct answer - regular pattern of up and down fluctuations - influenced by seasonal factors (month, quarter, weather, customs, etc.) - of a *fixed* and known period - typically occurs within a single year cyclical component - correct answer - repeating up and down movements that are not of a fixed period - typically multiple years in duration - may be due to casual or associative relationships - can be affected by business cycle, political, and economical factors random component - correct answer - erratic, non-systematic 'residual' fluctuations - due to random variation or unforeseen events - short duration and non-repeating basic steps in forecasting - correct answer 1. problem definition 2. gathering data 3. preliminary analysis 4. choosing and fitting models 5. using and evaluating a forecasting model prediction - correct answer typically uses explanatory variables to characterize expected outcomes forecast - correct answer primarily uses historical patterns and trends to characterize future performance; typically part of a bigger picture planning process forecasting approaches: quantitative methods - correct answer - naive approach - trend projection - linear regression - ARIMA models - exponential smoothing time series characteristics - correct answer collected over time, but must be made equally spaced, or indexed, before models can be used to quantify the systematic variation contained in it accumulation - correct answer the process of indexing or transforming transactional data into a time series missing values - correct answer should be represented in the data; can't have gaps embedded missing values - correct answer missing values at either the beginning or end; typically just trim off and reduce series imputing missing values - correct answer using forecasting to fill the missing values by using the mean, median, min, or max value, or using the previous or next v