Forecast Objectives Fin250f: Lecture 8.2 Spring 2010 Reading: Brooks, chapter
Outline Forecasting methodology and dangers Linear forecasts Forecast objectives
Forecasting Methodology Prediction about tomorrow In sample/Out of sample forecasts One step ahead/multi-step Direct/iterative Recursive/rolling windows
AR(1): Zero mean form
Forecasting the AR(1)
Forecasting the AR(1): Multiperiods
Forecasting an MA(1)
Exponential Smoothing Ad hoc forecasting tool Primitive (but maybe effective) Common in business forecasting environments
Exponential Smoothing
Forecast Objectives Mean Squared Error (MSE) Mean Absolute Error (MAE) Theil's U-statistic Sign predictions Trading profits
Mean Squared Error
Mean Absolute Error
Theil's U-statistic
Sign Prediction
Reminder on Central Limit Theorem
Binomial Test
Trading Profits
Comparisons How do you weight outliers? MSE: a lot Sign: Not much How does this connect to economics MSE: not much Sign: more Trading: a lot Mistakes connected to real losses Weights problems with large returns in a sensible way Actual trading is still more complex (costs, slippage)