Transcript Document

Forecasting

JY Le Boudec 1

Contents

1. What is forecasting ?

2. Linear Regression 3. Avoiding Overfitting 4. Differencing 5. ARMA models 6. Sparse ARMA models 7. Case Studies 2

1. What is forecasting ?

Assume you have been able to define the nature of the load It remains to have an idea about its intensity It is impossible to forecast without error The good engineer should Forecast

what can be forecast

Give

uncertainty

intervals The rest is outside our control 3

4

2. Linear Regression

Simple, for simple cases Based on extrapolating the explanatory variables 5

6

7

8

Estimation and Forecasting

In practice we estimate  from y, …, y t When computing the forecast, we pretend  estimation error is known, and thus make an It is hoped that the estimation error is much less than the confidence interval for forecast In the case of linear regression, the theorem gives the global error exactly In general, we won’t have this luxury 9

10

We saw this already

A case where estimation error versus prediction uncertainty can be quantified Prediction interval if model is known Prediction interval accounting for estimation (t = 100 observed points) 11

3. The Overfitting Problem

The best model is not necessarily the one that fits best 12

Prediction for the better model

This is the overfitting problem 13

How to avoid overfitting

Method 1: use of test data Method 2: information criterion 14

15

16

Best Model for Internet Data, polynomial of degree up to 2

17

d = 1

18

Best Model for Internet Data, polynomial of degree up to 10

19

4. Differencing the Data

20

21

22

Point Predictions from Differenced Data

23

Background On Filters (Appendix B)

We need to understand how to use discrete filters.

Example: write the Matlab command for 24

25

A simple filter

Q: compute X back from Y 26

27

28

29

Impulse Response

30

31

32

A filter with stable inverse

33

How is this prediction done ?

This is all very intuitive 34

35

Prediction assuming differenced data is iid

36

Prediction Intervals

A prediction without prediction intervals is only a small part of the story The financial crisis might have been avoided if investors had been aware of prediction intervals 37

38

39

Compare the Two

Linear Regression with 3 parameters + variance Assuming differenced data is iid 40

41

5. Using ARMA Models

When the differenced data appears stationary but not iid 42

Test of iid-ness

43

44

ARMA Process

45

46

ARMA Processes are Gaussian (non iid)

47

48

49

AR I MA Process

50

Fitting an ARMA Process

Called the Box-Jenkins method Difference the data until stationary Examine ACF to get a feeling of order (p,q) Fit an ARMA model using maximum likelihood 51

Fitting an ARIMA Model

Apply Scientific Method 1. make stationary and normal (how ?) 2. bound orders p,q 3. fit an ARMA model to Y t  i.e. Y t  » ARMA 4. compute residuals and verify white noise and normal Fitting an ARMA model Pb is : given orders p,q given (x 1 , …x likelihood n ) (transformed data) compute the parameters of an ARMA (p,q) model that maximizes the

Q:What are the parameters ?

A: the mean

, the polynomial coefficients variance

2

k and

k , the noise

52

This is a non-linear optimization problem

Maximizing the likelihood is a non-linear optimization problems Usually solved by iterative, heuristic algorithms, may converge to a local maximum may not converge Some simple, non MLE, heuristics exist for AR or MA models Ex: fit the AR model that has the same theoretical ACF as the sample ACF Common practice is to bootstrap the optimization procedure by starting with a “best guess” AR or MA fit, using heuristic above 53

Fitting ARMA Model is Same as Minimizing One-Step ahead prediction error

54

Best Model Order

55

Check the Residuals

56

Example

57

58

Forecasting with ARMA

Assume Y t is fitted to an ARMA process The prediction problem is: given Y 1 =y 1 ,…,Y t =y t find the conditional distribution of Y t+h We know it is normal, with a mean that depends on (y 1 ,…,y t ) and a variance that depends only on the fitted parameters of the ARMA process There are many ways to compute this; it is readily done by Matlab 59

60

61

Forecasting Formulae for ARIMA

Y = original data X = differenced data, fitted to an ARMA model 1.

2.

3.

Obtain point prediction for X using what we just saw Apply Proposition 6.4.1 to obtain point prediction for Y Apply formula for prediction interval There are several other methods, but they may have numerical problems. See comments in the lecture notes after prop 6.5.2 62

63

Improve Confidence Interval If Residuals are not Gaussian (but appear to be iid)

Assume residuals are not gaussian but are iid How can we get confidence intervals ?

Bootstrap by sampling from residuals 64

65

With gaussian assumption With bootstrap from residuals 66

6. Sparse ARMA Models

Problem: avoid many parameters when the degree of the A and C polynomials is high, as in the previous example Based on heuristics Multiplicative ARIMA, constrained ARIMA Holt Winters 67

68

Holt Winters Model 1: EWMA

69

70

EWMA is OK when there is no trend and no periodicity

71

72

73

74

75

76

77

(corrected or not) Constrained ARIMA Sparse models give less accurate predictions but have much fewer parameters and are simple to fit.

78

7. Case Studies

79

80

81

82

83

84

85

86

87

h = 1 88

89

h = 2 90

91

log h = 1 92

Conclusion

Forecasting is useful when savings matter; for example Save money on server space rental Save energy Capturing determinism is perhaps most important and easiest Prediction intervals are useful to avoid gross mistakes Re-scaling the data may help … à vous de jouer.

93

94