TS_Fall_2011_Lecture_3

TS_Fall_2011_Lecture_3 - FIN 271 FINANCIAL MODELING AND...

Info iconThis preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon
FIN 271 FINANCIAL MODELING AND ECONOMETRICS LECTURE SET 3 TIME SERIES MODELING REFIK SOYER THE GEORGE WASHINGTON UNIVERSITY
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 Forecasting from AR(1) Model Once the coefficients are estimated, we can forecast the future values of the TS . ] t Let (h) denote the forecast of at time n after observing , , . . ., . ] ] ] ] ] s n n h 1 2 n ] ´ s n (h) h-step ahead forecast at time n h lead time, n origin (of the forecast) ´ ´ For example, (1) is the one-step ahead forecast at time n. ] s n Consider the AR(1) process: ] œ G  ] t 1 t 1 t 9 % We are interested in forecasting ] n 1 ] œ G  ]  n 1 1 n n 1 9 % Given , , . . ., , one-step ahead forecast at time t ] ] ] 1 2 n ] Ð Ñ œ G  ] s n 1 n 1 9 1
Background image of page 2
3 Note that expected value of given past observations is 0 % n 1 Þ From the above we can write % n 1 n 1 n œ ]  ] Ð Ñ s 1 Ê is the one-step ahead forecast error ! % n 1 What about forecast for ? ] n 2 ] œ G  ] n 2 1 n 1 n 2 9 % ] Ð Ñ œ G  ] Ð Ñ s s n 1 n 2 1 9 ] Ð Ñ œ G  ÐG  ] Ñ œ Ð  ÑG  ] s n 1 1 n 1 n 2 1 2 1 9 9 9 9 In general for AR 1 Process: Ð Ñ ] Ð Ñ œ Ð   á  ÑG  ] s n 1 n 1 1 h 1 h h 1 9 9 9 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
4 MEAN REVERSION OF FORECASTS What happens when h gets large ? In the above 1 is a geometric series thus it converges to Ð   á  Ñ 9 9 1 1 h 1 1 and thus ÎÐ"  Ñ 9 1 ] Ð Ñ Ä œ s G Ð  Ñ n 1 h . 1 9 . As the lead time increases the forecasts get closer to the mean and eventually the forecasts revert to the mean. This is a property of stationary time series models and will not apply to nonstationary series. See SAS example on price of the crude oil. 3
Background image of page 4
Analysis of the Price of Crude Oil Data * AVERAGE PRICE OF CRUDE OIL AT THE WELL JAN74-DEC82 INDEX=100 IN 1967; DATA OIL; INFILE 'C:\TSDATA\Crudeoil.txt'; INPUT PGAS PCRUDE STOCKGAS GASCONS STOCKOIL; PROC ARIMA DATA=OIL; IDENTIFY VAR=PCRUDE; RUN; Name of variable = PCRUDE. Mean of working series = 4173.5 Standard deviation = 2131.784 Number of observations = 108 Autocorrelations Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 1 0 4544501 1.00000 | |********************| 1 4458346 0.98104 | . |********************| 2 4356540 0.95864 | . |******************* | 3 4244987 0.93409 | . |******************* | 4 4130834 0.90897 | . |****************** | 5 4014015 0.88327 | . |****************** | 6 3896140 0.85733 | . |***************** | 7 3780944 0.83198 | . |***************** | 8 3662695 0.80596 | . |**************** | 9 3541294 0.77925 | . |**************** | 10 3411041 0.75059 | . |*************** | 11 3267858 0.71908 | . |**************. | 12 3113620 0.68514 | . |************** . | 13 2957142 0.65071 | . |************* . | 14 2797785 0.61564 | . |************ . | 15 2638093 0.58050 | . |************ . | 16 2474790 0.54457 | . |*********** . | 17 2307640 0.50779 | . |********** . | 18 2139269 0.47074 | . |********* . | 19 1961795 0.43169 | . |********* . | 20 1776540 0.39092 | . |******** . | 1 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 6
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/29/2012 for the course FINA 6271 taught by Professor Phillipwirtz,refiksoyer during the Fall '11 term at GWU.

Page1 / 40

TS_Fall_2011_Lecture_3 - FIN 271 FINANCIAL MODELING AND...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online