econ 240 c lecture 16. 2 outline w project i w arch-m models w granger causality w simultaneity w...

90
Econ 240 C Lecture 16

Post on 22-Dec-2015

213 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

Econ 240 C

Lecture 16

Page 2: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

2

Outline Project I ARCH-M Models Granger Causality Simultaneity VAR models

Page 3: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

3Project I Models for dduration Models for dlnduration Seasonality Conditional heteroskedasticity

Page 4: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

4

Models for ∆duration

Ufook Sahillioghlu• Ar(1) ar(2) ar(4) ar(5) ar(6) ma(7) ma(24) ma(36)

Tom Bruister• Ar(1) ar(2) ar(24) ma(1) ma(4)

Jesse Smith• Ar(1) ar(4) ar(24) ar(36)

Page 5: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

5Models for ∆lnduration Jonathan Hester

• Ar(1) ma(1) ma(2) ma(3) Ashley Hedberg

• Ar(1) ar(2) ma(1) ma(2) Jonathan Liu

• Ar(1) ar(2) ar(4) ar(5) ar(6) ma(7) ma(24) ma(36) Yana Ten

• Ma(1) ma(4) ar(24) ar(36) Jeff Ahlvin

• Ma(1) ma(2) ma(3) sma(24)

Page 6: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

6

Page 7: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

7

Page 8: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

8

Page 9: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

9

Page 10: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

10

Page 11: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

11

Page 12: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

12

Conditional Variance, h

Page 13: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

13

Part I. ARCH-M Modeks

In an ARCH-M model, the conditional variance is introduced into the equation for the mean as an explanatory variable.

ARCH-M is often used in financial models

Page 14: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

14Net return to an asset model Net return to an asset: y(t)

• y(t) = u(t) + e(t)• where u(t) is is the expected risk premium• e(t) is the asset specific shock

the expected risk premium: u(t)• u(t) = a + b*h(t)• h(t) is the conditional variance

Combining, we obtain:• y(t) = a + b*h(t) +e(t)

Page 15: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

15Northern Telecom And Toronto Stock Exchange

Nortel and TSE monthly rates of return on the stock and the market, respectively

Keller and Warrack, 6th ed. Xm 18-06 data file

We used a similar file for GE and S_P_Index01 last Fall in Lab 6 of Econ 240A

Page 16: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

16

Page 17: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

17Returns Generating Model, Variables Not Net of Risk Free

Page 18: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

18

Page 19: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

19Diagnostics: Correlogram of the Residuals

Page 20: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

20Diagnostics: Correlogram of Residuals Squared

Page 21: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

21

Page 22: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

22Try Estimating An ARCH-

GARCH Model

Page 23: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

23

Page 24: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

24Try Adding the Conditional Variance to the Returns Model PROCS: Make GARCH variance series:

GARCH01 series

Page 25: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

25Conditional Variance Does Not Explain Nortel Return

Page 26: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

26

Page 27: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

27OLS ARCH-M

Page 28: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

28

Estimate ARCH-M Model

Page 29: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

29Estimating Arch-M in Eviews with GARCH

Page 30: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

30

Page 31: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

31

Page 32: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

32Three-Mile Island

Page 33: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

33

Page 34: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

34

Page 35: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

35

Page 36: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

36

Event: March 28, 1979

Page 37: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

37

Page 38: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

38

Page 39: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

39Garch01 as a Geometric Lag of GPUnet

Garch01(t) = {b/[1-(1-b)z]} zm gpunet(t) Garch01(t) = (1-b) garch01(t-1) + b zm gpunet

Page 40: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

40

Page 41: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

41

Part II. Granger Causality

Granger causality is based on the notion of the past causing the present

example: Lab six, Index of Consumer Sentiment January 1978 - March 2003 and S&P500 total return, montly January 1970 - March 2003

Page 42: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

42Consumer Sentiment and SP 500 Total Return

Page 43: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

43

Time Series are Evolutionary

Take logarithms and first difference

Page 44: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

44

Page 45: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

45

Page 46: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

46

Dlncon’s dependence on its past

dlncon(t) = a + b*dlncon(t-1) + c*dlncon(t-2) + d*dlncon(t-3) + resid(t)

Page 47: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

47

Page 48: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

48Dlncon’s dependence on its past and dlnsp’s past

dlncon(t) = a + b*dlncon(t-1) + c*dlncon(t-2) + d*dlncon(t-3) + e*dlnsp(t-1) + f*dlnsp(t-2) + g* dlnsp(t-3) + resid(t)

Page 49: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

49

Page 50: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

Do lagged dlnsp terms add to the explained variance?

F3, 292 = {[ssr(eq. 1) - ssr(eq. 2)]/3}/[ssr(eq. 2)/n-7]

F3, 292 = {[0.642038 - 0.575445]/3}/0.575445/292

F3, 292 = 11.26

critical value at 5% level for F(3, infinity) = 2.60

Page 51: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

51

Causality goes from dlnsp to dlncon

EVIEWS Granger Causality Test• open dlncon and dlnsp• go to VIEW menu and select Granger Causality• choose the number of lags

Page 52: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

52

Page 53: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

53Does the causality go the other way, from dlncon to dlnsp? dlnsp(t) = a + b*dlnsp(t-1) + c*dlnsp(t-2) +

d* dlnsp(t-3) + resid(t)

Page 54: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

54

Page 55: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

55Dlnsp’s dependence on its past and dlncon’s past dlnsp(t) = a + b*dlnsp(t-1) + c*dlnsp(t-2) +

d* dlnsp(t-3) + e*dlncon(t-1) + f*dlncon(t-2) + g*dlncon(t-3) + resid(t)

Page 56: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

56

Page 57: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

Do lagged dlncon terms add to the explained variance?

F3, 292 = {[ssr(eq. 1) - ssr(eq. 2)]/3}/[ssr(eq. 2)/n-7]

F3, 292 = {[0.609075 - 0.606715]/3}/0.606715/292

F3, 292 = 0.379

critical value at 5% level for F(3, infinity) = 2.60

Page 58: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

58

Page 59: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

59Granger Causality and Cross-Correlation

One-way causality from dlnsp to dlncon reinforces the results inferred from the cross-correlation function

Page 60: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

60

Page 61: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

61Part III. Simultaneous Equations

and Identification Lecture 2, Section I Econ 240C Spring

2005 Sometimes in microeconomics it is possible

to identify, for example, supply and demand, if there are exogenous variables that cause the curves to shift, such as weather (rainfall) for supply and income for demand

Page 62: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

62

Demand: p = a - b*q +c*y + ep

Page 63: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

63

demand

price

quantity

Dependence of price on quantity and vice versa

Page 64: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

64

demand

price

quantity

Shift in demand with increased income

Page 65: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

65

Supply: q= d + e*p + f*w + eq

Page 66: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

66

price

quantity

supply

Dependence of price on quantity and vice versa

Page 67: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

67

Simultaneity

There are two relations that show the dependence of price on quantity and vice versa• demand: p = a - b*q +c*y + ep

• supply: q= d + e*p + f*w + eq

Page 68: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

68

Endogeneity

Price and quantity are mutually determined by demand and supply, i.e. determined internal to the model, hence the name endogenous variables

income and weather are presumed determined outside the model, hence the name exogenous variables

Page 69: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

69

price

quantity

supply

Shift in supply with increased rainfall

Page 70: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

70

Identification

Suppose income is increasing but weather is staying the same

Page 71: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

71

demand

price

quantity

Shift in demand with increased income, may trace outi.e. identify or reveal the demand curve

supply

Page 72: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

72

price

quantity

Shift in demand with increased income, may trace outi.e. identify or reveal the supply curve

supply

Page 73: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

73

Identification

Suppose rainfall is increasing but income is staying the same

Page 74: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

74

price

quantity

supply

Shift in supply with increased rainfall may trace out, i.e. identify or reveal the demand curve

demand

Page 75: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

75

price

quantity

Shift in supply with increased rainfall may trace out, i.e. identify or reveal the demand curve

demand

Page 76: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

76

Identification

Suppose both income and weather are changing

Page 77: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

77

price

quantity

supply

Shift in supply with increased rainfall and shift in demandwith increased income

demand

Page 78: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

78

price

quantity

Shift in supply with increased rainfall and shift in demandwith increased income. You observe price and quantity

Page 79: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

79

Identification

All may not be lost, if parameters of interest such as a and b can be determined from the dependence of price on income and weather and the dependence of quantity on income and weather then the demand model can be identified and so can supply

Page 80: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

The Reduced Form for p~(y,w)

demand: p = a - b*q +c*y + ep

supply: q= d + e*p + f*w + eq

Substitute expression for q into the demand equation and solve for p

p = a - b*[d + e*p + f*w + eq] +c*y + ep

p = a - b*d - b*e*p - b*f*w - b* eq + c*y + ep

p[1 + b*e] = [a - b*d ] - b*f*w + c*y + [ep - b* eq ]

divide through by [1 + b*e]

Page 81: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

The reduced form for q~y,w

demand: p = a - b*q +c*y + ep

supply: q= d + e*p + f*w + eq

Substitute expression for p into the supply equation and solve for q

supply: q= d + e*[a - b*q +c*y + ep] + f*w + eq

q = d + e*a - e*b*q + e*c*y +e* ep + f*w + eq

q[1 + e*b] = [d + e*a] + e*c*y + f*w + [eq + e* ep]

divide through by [1 + e*b]

Page 82: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

Working back to the structural parameters

Note: the coefficient on income, y, in the equation for q, divided by the coefficient on income in the equation for p equals e, the slope of the supply equation• e*c/[1+e*b]÷ c/[1+e*b] = e

Note: the coefficient on weather in the equation f for p, divided by the coefficient on weather in the equation for q equals -b, the slope of the demand equation

Page 83: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

This process is called identification

From these estimates of e and b we can calculate [1 +b*e] and obtain c from the coefficient on income in the price equation and obtain f from the coefficient on weather in the quantity equation

it is possible to obtain a and d as well

Page 84: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

84

Vector Autoregression (VAR)

Simultaneity is also a problem in macro economics and is often complicated by the fact that there are not obvious exogenous variables like income and weather to save the day

As John Muir said, “everything in the universe is connected to everything else”

Page 85: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

85VAR One possibility is to take advantage of the

dependence of a macro variable on its own past and the past of other endogenous variables. That is the approach of VAR, similar to the specification of Granger Causality tests

One difficulty is identification, working back from the equations we estimate, unlike the demand and supply example above

We miss our equation specific exogenous variables, income and weather

Page 86: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

Primitive VAR

(1)y(t) = w(t) + y(t-1) +

w(t-1) + x(t) + ey(t)

(2) w(t) = y(t) + y(t-1) +

w(t-1) + x(t) + ew(t)

Page 87: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

87

Standard VAR

Eliminate dependence of y(t) on contemporaneous w(t) by substituting for w(t) in equation (1) from its expression (RHS) in equation 2

Page 88: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

1. y(t) = w(t) + y(t-1) + w(t-1) + x(t) + ey(t)

1’. y(t) = y(t) + y(t-1) + w(t-1) + x(t) + ew(t)] + y(t-1) + w(t-1) + x(t) + ey(t)

1’. y(t) y(t) = [+ y(t-1) + w(t-1) + x(t) + ew(t)] + y(t-1) + w(t-1) + x(t) + ey(t)

Page 89: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

Standard VAR (1’) y(t) = (/(1- ) +[ (+

)/(1- )] y(t-1) + [ (+ )/(1- )] w(t-1) + [(+ (1- )] x(t) + (ey(t) + ew(t))/(1- )

in the this standard VAR, y(t) depends only on lagged y(t-1) and w(t-1), called predetermined variables, i.e. determined in the past

Note: the error term in Eq. 1’, (ey(t) + ew(t))/(1- ), depends upon both the pure shock to y, ey(t) , and the pure shock to w, ew

Page 90: Econ 240 C Lecture 16. 2 Outline w Project I w ARCH-M Models w Granger Causality w Simultaneity w VAR models

Standard VAR (1’) y(t) = (/(1- ) +[ (+ )/(1-

)] y(t-1) + [ (+ )/(1- )] w(t-1) + [(+ (1- )] x(t) + (ey(t) + ew(t))/(1- )

(2’) w(t) = (/(1- ) +[(+ )/(1- )] y(t-1) + [ (+ )/(1- )] w(t-1) + [(+ (1- )] x(t) + (ey(t) + ew(t))/(1- )

Note: it is not possible to go from the standard VAR to the primitive VAR by taking ratios of estimated parameters in the standard VAR