Mod 3C Time Series & Forecasting Flashcards
What is the difference between qualitative and quantitative forecasting methods?
Qualitative forecasting methods rely on expert judgement to predict future scenarios without requiring historical data, while quantitative methods use historical data assuming past performance predicts future outcomes.
What is a time series in the context of forecasting?
A time series is a sequence of observations on a variable measured at successive points in time, used to uncover patterns and predict future values.
What are the key patterns to identify in time series analysis?
Horizontal (data fluctuates around a constant mean), Seasonal (occurs over periods), and Cyclical (alternating sequence lasting more than a year).
What are the main metrics to measure forecast accuracy?
Mean Forecast Error (MFE), Mean Absolute Error (MAE), Mean Squared Error (MSE), and Mean Absolute Percentage Error (MAPE).
How does the moving averages method work?
It uses the average of the most recent ‘k’ data values in the time series as the forecast for the next period.
What is exponential smoothing in time series forecasting?
It uses a weighted average of past time series values as a forecast, giving more weight to recent observations.
What is linear trend projection in time series forecasting?
A regression analysis method using time (independent variable) to forecast the future values of the time series (dependent variable).
What are autoregressive models in forecasting?
Regression models where the independent variables are previous values of the time series itself.
How can seasonality be modeled in time series analysis?
By treating seasons as dummy variables, with K-1 dummy variables for a variable with K levels.
What factors are considered in selecting the best model for time series forecasting?
Ability to identify and use relevant independent variables, visual inspection for seasonality and trends, and minimizing forecast errors in validation sets.
