Presentation is loading. Please wait.

Presentation is loading. Please wait.

1 Project I Econ 240c Spring 2006. 2 Issues  Parsimonious models  2006: March or April 9.3 wks or 8.9 wks  Trend  Residual seasonality  Forecasts:

Similar presentations


Presentation on theme: "1 Project I Econ 240c Spring 2006. 2 Issues  Parsimonious models  2006: March or April 9.3 wks or 8.9 wks  Trend  Residual seasonality  Forecasts:"— Presentation transcript:

1 1 Project I Econ 240c Spring 2006

2 2 Issues  Parsimonious models  2006: March or April 9.3 wks or 8.9 wks  Trend  Residual seasonality  Forecasts: sharp peaks or broad peaks?  Model selection  The labor market  Trend  Broad peaks

3 3 ∆ duration  : ar(1) ar(4) ar(24) ar(36)  : ma(1) ma(4) ma(24) ma(36)  : ar(1) ar(2) ma(1) ma(2)

4 4 ∆ lnduration  : ma(1) ma(4) sma(24) sma(36)  : ma(1) ma(4) ar(24) ar(36)  : ar(1) ar(2) ma(1) ma(2)  : ar(1) ma(1) ma(2) ma(3)

5 5 Outline  Duration Model: trend and arma error, p.6  Dduration model: arma(2,2), p. 29

6 6 Duration in Levels  Trend  Duration = a + b*t +arma error  Identification  Estimation  Model Verification  Within Sample Forecasting, a Test of the Model  Forecasts

7 7

8 8

9 9

10 10

11 11

12 12

13 13

14 14

15 15

16 16

17 17 Conditional Heteroskedasticity?

18 18

19 19

20 20 Within Sample Test

21 21

22 22

23 23 Quick Menu, Show

24 24

25 25 Forecast: 2006.05-2007.12

26 26

27 27 Quick Menu, Show

28 28

29 29 Monthly Change in Duration

30 30

31 31

32 32

33 33

34 34

35 35

36 36

37 37

38 38 Dduration: ar(1) ar(2), ma(1)

39 39 Dduration: ar(1) ar(2), ma(2)

40 40Model

41 41 Diagnostics

42 42

43 43

44 44

45 45

46 46

47 47 Within Sample Forecast

48 48

49 49

50 50 Quick menu, show

51 51 Within Sample Forecast

52 52 Out of Sample Forecast

53 53

54 54

55 55

56 56

57 57

58 58

59 59

60 60

61 61 Fractional Change in Duration

62 62

63 63 Identification

64 64 Model

65 65 Model verification

66 66

67 67

68 68


Download ppt "1 Project I Econ 240c Spring 2006. 2 Issues  Parsimonious models  2006: March or April 9.3 wks or 8.9 wks  Trend  Residual seasonality  Forecasts:"

Similar presentations


Ads by Google