11. Time Series Analysis
Basic Regression
Read Wooldridge (2013), Chapter 10
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat
Outline
I. The Nature of Time Series Data II. Examples of Time Series Models III. Finite Sample Properties of OLS
IV. Functional Form, Dummy Variables, and Index Numbers
I. Nature II. Examples III. Properties IV. Index 2
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat
I. The Nature of Time Series Data
• Time series
A time series data set consists of observations on a variable or several variables over time.
eg. GDP, money supply, interest rate.
• Cross section series
A cross sectional data set is a data set collected from a population at a given point in time.
I. Nature II. Examples III. Properties IV. Index 3
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat I. The Nature of Time Series Data
Differences between Time Series and Cross Section
(1) Ordering
Cross section: Ordering usually is not important.
Time series: A data set comes with a temporal ordering (2) Random sample
Cross section: A random sample is drawn from the population. Each observation is randomly drawn (MLR.2).
Time series: An observation is an outcome of random variables.
I. Nature II. Examples III. Properties IV. Index 4
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat I. The Nature of Time Series Data
Terms:
• Stochastic – random
• Realization – observation
• Formally, say observation 1960‐2003.
A sequence of random variables indexed by time is called a stochastic process or a time series process.
– At a point in time, we obtain a single realization of the stochastic process. Note that we cannot go back in time.
I. Nature II. Examples III. Properties IV. Index 5
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat I. The Nature of Time Series Data
II. Examples of Time Series Regression
• Static and Finite Distributed Lag Models
• Static Model
y t = 0 + 1 z t + u t t = 1, 2,…,n – A static model is a model that relates y to z using the
same time period.
– 1 : An immediate effect of z on y.
I. Nature II. Examples III. Properties IV. Index 6
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat II. Examples of Time Series Regression
Examples of Time Series Regression
• Example : Phillips Curve inf t = 0 + 1 unem t + u t
inf t : annual inflation rate (%) unem t : unemployment rate (%)
• = 1.423+ .468unem (s.e) (1.72) (.289)
[t] [.828] {1.617}
n=49 R 2 =.053 R 2 bar=.032 – How to define 1 ?
I. Nature II. Examples III. Properties IV. Index 7
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat II. Examples of Time Series Regression
Finite Distributed Lag (FDL) Models:
Effect of one or more variables with a lag on y
• Example: Effect of the growth in money supply on economic growth in Thailand over 1993‐2002 (quarterly data)
ggdp: economic growth (percent) gm1: money supply growth (percent) ggdp
t=
0+
0gm1
t+
1gm1
t‐1+
2gm1
t‐2+ u
t• Interpret coefficients: let t = 0
0: “this quarter” effect of MS growth on “this quarter” ggdp.
1: “last quarter” effect of MS growth on “this quarter” ggdp
2: “two‐quarters ago” effect of MS growth on “this quarter” ggdp.
I. Nature II. Examples III. Properties IV. Index 8
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat II. Examples of Time Series Regression
Example: effect of gm1 on ggdp y t = 0 + 0 z t + 1 z t‐1 + 2 z t‐2 + u t
t = .001 + .319gm1 t +.121gm1 t‐1 ‐.126gm1 t‐2 p‐value {.88} {.0005} {.105} {.1359}
n=39, R 2 =.5088 F=12.08 {p‐value=.000014}
Interpretation: Economic Significance.
1) Interpret 2) Interpret 3) Interpret
I. Nature II. Examples III. Properties IV. Index 9
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat II. Examples of Time Series Regression
A General form – a FDL of order two y t = 0 + 0 z t + 1 z t‐1 + 2 z t‐2 + u t
Interpret coefficients:
1) 0 : the immediate change in y due to the one‐unit increase in z at time t
– 0 is called the impact propensity or impact multiplier.
2) 0 + 1 + 2 : the long‐run change in y given a permanent increase in z
– 0 + 1 + 2 is called the long run propensity (LRP) or long‐run multiplier.
I. Nature II. Examples III. Properties IV. Index 10
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat II. Examples of Time Series Regression
A temporal increase in z: j
y t = 0 + 0 z t + 1 z t‐1 + 2 z t‐2 + u t
“temporal” means lasting only for a time.
• Assumptions:
1) Before time t: z is a constant (c) 2) At time t, z increases one unit to c+1
3) After time t, z reverts back to its previous level, c
• Interpretation:
–
0= y
t– y
t‐1: immediate change in y due to the one‐unit increase in z at time t
–
1= y
t+1– y
t‐1: the change in y one period after the increase in z.
–
2= y
t+2– y
t‐1: the change in y two periods after the increase in z.
I. Nature II. Examples III. Properties IV. Index 11
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat II. Examples of Time Series Regression
A permanent increase in z: 0 + 1 +
2y t = 0 + 0 z t + 1 z t‐1 + 2 z t‐2 + u t
• Assumptions:
1) Before time t, z is a constant (c).
2) At time t, z increases one unit permanently to c+1.
• Interpretation: With the permanent increase in z
0= y
t– y
t‐1: immediate change
0+
1= y
t+1– y
t‐1: increase in y after one period.
0+
1+
2= y
t+2– y
t‐1: increase in y after two periods.
•
0+
1+
2is the LR change in y given a permanent increase in z in the FDL model of order 2.
I. Nature II. Examples III. Properties IV. Index 12
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat II. Examples of Time Series Regression
Example: effect of gm1 on ggdp y t = 0 + 0 z t + 1 z t‐1 + 2 z t‐2 + u t
t = .001 + .319gm1 t +.121gm1 t‐1 ‐.126gm1 t‐2 p‐value {.88} {.0005} {.105} {.1359}
n=39, R 2 =.5088 F=12.08 {p‐value=.000014}
Interpretation: Economic and Statistical Significance?
– Impact multiplier = ? and Test?
– Long run multiplier = ? and Test?
• Graph: a lag distribution with two nonzero lags.
I. Nature II. Examples III. Properties IV. Index 13
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat II. Examples of Time Series Regression
Dependent Variable: GGDP Method: Least Squares Sample(adjusted): 1993:2 2002:4
Included observations: 39 after adjusting endpoints
Variable Coefficient Std. Error t-Statistic Prob.
C 0.001303 0.006156 0.211701 0.8336
GM1 0.319434 0.082622 3.866224 0.0005
GM1(-1) 0.120663 0.072506 1.664191 0.105
GM1(-2) -0.12604 0.082562 -1.52659 0.1359
R-squared 0.50881 Mean dependent var 0.008038
Adjusted R-squared 0.466708 S.D. dependent var 0.040941
S.E. of regression 0.029898 Akaike info criterion -4.08516
Sum squared resid 0.031286 Schwarz criterion -3.91453
Log likelihood 83.66052 F-statistic 12.08517
Durbin-Watson stat 2.033647 Prob(F-statistic) 0.000014
Effect of Monetary Policy on Economic Growth
I. Nature II. Examples III. Properties IV. Index 14
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat II. Examples of Time Series Regression
III. Finite Sample Properties of OLS under classical Assumptions
TS.1: Linear in parameters.
Given stochastic process {(x
t1, x
t2,…,x
tk, y
t): t = 1, 2, .., n}
y
t=
0+
1x
t1+ … +
kx
tk+ u
twhere {u
t: t = 1, 2, …., n} is the sequence of errors.
x
tj: t denotes the time period j indicates one of the k variables
Unbiasedness of OLS: TS.1‐TS.3.
I. Nature II. Examples III. Properties IV. Index 15
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
TS.2: No perfect collinearity
No independent variable is constant or a perfect linear combination of the others.
TS.3 : Zero Conditional Mean E(u t |X) = 0
where X is an array with n rows and k columns.
I. Nature II. Examples III. Properties IV. Index 16
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
Example: y t =GDP; x t1 =M1 t x t2 =GE t (gov’t spending); n=44, k=2
year y
t= gdp x
t1= M1 x
t2= GE error
1960 gdp
60M1
60GE
60u
601961 gdp
61M1
61GE
61u
611962 gdp
62M1
62GE
62u
62.. .. .. .. ..
2001 gdp
01M1
01GE
01u
012002 gdp
02M1
02GE
02u
022003 gdp
03M1
03GE
03u
03u
tis uncorrelated with each explanatory variable in every time period. We say that x
tjare strictly exogenous.
I. Nature II. Examples III. Properties IV. Index 17
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
Theorem 3.1: Unbiasedness of OLS
If TS.1 – TS.3 hold, then the OLS estimators are unbiased, conditional in X, i.e.,
E( ) = j for j = 1, ..., k
I. Nature II. Examples III. Properties IV. Index 18
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
TS. 3' implies that u t is uncorrelated with regressors dated at time t, (TS.3' next chapter)
E(u t |x t1 ,…,x tk ) = E(u t |x) = 0
We say x tj are contemporaneously exogenous.
• Why don’t we assume E(u i |X)=0 or strict exogeneity in the cross sectional analysis?
• Random sampling: u i is automatically independent of the explanatory variables other than i.
year y
t= gdp x
t1= M1 x
t2= GE error
2001 gdp
01M1
01GE
01u
01I. Nature II. Examples III. Properties IV. Index 19
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
Example: Strict Exogenity Assumption Model: y t = 0 + 1 z t + u t
ggdp t = 0 + 1 gm1 t + u t TS. 3 requires that
(1) u t and z t are uncorrelated
(2) u t is also unrelated with past and future values of z;
that is, z have no lagged effect on y.
(3) A subtle point : the changes in error term today cannot cause future changes in z. ( u t y t z t+1 )
( u t ggdp t gm1 t+1 )
• This rules out feedback from y on future values of z.
I. Nature II. Examples III. Properties IV. Index 20
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
Example: Murder rate equation mrdrte t = 0 + 1 polpc t + u t
mrdrte: murders per 10,000 people polpc: number of police in the force
• Two implications:
(1) TS.3 implies that u t is uncorrelated with polpc in all time period.
Violation: Higher u 0 may lead to larger polpc 1 force.
(2) Explanatory variables that are strictly exogeneous cannot react to what has happened to y in the past.
year=t y t =mrdrte t z t =polpc t u t =error
2000 10,000 100 u 0
2001 …. 150 ..
Violation: high u 0 high mrdrt 0 high polpc 1
? ?
I. Nature II. Examples III. Properties IV. Index 21
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
Cases: Agricultural Production, rainfall and labor input
Example : Agricultural Production (y)
Case 1: y=output z=rainfall u t =error
t=0 y 2000 u 2000
t=1 z 2001
Case 2: y=output z=labor input u t =error
t=0 z 2000 u 2000
t=1 z 2001
?
?
?
?
I. Nature II. Examples III. Properties IV. Index 22
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
TS.4: Homoskedasticity
VAR(u t |X) = VAR(u t ) = 2 t = 1 ,…, n
• Example: Effect on Treasury bills i3 t = 0 + 1 inf t + 2 def t + u t
def t : federal deficit as a percentage of GDP.
• TS. 4 requires that unobservables affecting interest rates have a constant variance over time.
• Violation: Variability of interest rates depends on the level of inflation or relative size of the deficits.
I. Nature II. Examples III. Properties IV. Index 23
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
TS.5: No Serial Correlation
Corr(u s ,u t |X) = 0 for all t s
Example: No Serial Correlation
y t = 0 + 1 inf t + 2 def t + u t
• When u t‐1 > 0 then, on average, the error in the next period u t is positive, or Corr(u t , u t‐1 ) > 0.
– This problem is called serial correlation or auto correlation.
• This implies that if interest rate is high in this period, it will be high in the next period: a violation of TS.5.
I. Nature II. Examples III. Properties IV. Index 24
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
25
Positive Serial Correlation
Negative Serial Correlation
Serial Correlation
I. Nature II. Examples III. Properties IV. Index 25
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
• Theorem 10.2: OLS Sampling Variances If TS.1 – TS.5 hold, then
Var( X ) =
j = 1, …, k, where SST j is the total sum of squares of x tj and R j 2 is the regression of x j on the other regressors.
I. Nature II. Examples III. Properties IV. Index 26
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
• Theorem 10.4: Gauss Markov Theorem
If TS.1‐TS.5 (Gauss Markov assumptions) hold, then conditional on X, the OLS estimators are the best linear unbiased estimators (BLUE).
I. Nature II. Examples III. Properties IV. Index 27
• Theorem 10.3: Unbiased estimator of 2
If TS.1 – TS.5 hold, then the unbiased estimator of 2 is
2 =
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
Inference under the CLM Assumptions
• Assumption TS.6: Normality assumption The errors u t are independent of X and
u t N(0, 2 )
That is, u t is independently and identically distributed (i.i.d) as Normal (0, 2 ).
I. Nature II. Examples III. Properties IV. Index 28
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
Inference under the CLM Assumptions
Theorem 10.5: Normal Sampling Distribution If TS.1‐TS.6 (CLM assumptions) hold, then (1) is normally distributed,
(2) t and F statistics have t and F distributions, respectively, under the null hypothesis, and
(3) the usual construction of confidence interval is valid.
I. Nature II. Examples III. Properties IV. Index 29
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
Example: Effects of inflation and deficits on interest rate in the United States over 1948‐66.
i3: the three‐month T‐bill rate
def: federal deficit as a percentage of GDP 3 t = 1.25 + 0.613inf t + 0.700def t
s.e. (0.44) (0.076) (0.118) t [2.84] [8.06] [5.93]
n = 49 R 2 = 0.697 R 2 bar= 0.683
• Interpret: Economic and Statistical Significance
1) Test inf and def at the 5% level.
2) Interpret the coefficient on inf
3) Interpret the coefficient on def
I. Nature II. Examples III. Properties IV. Index 30
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat III. Finite Sample Properties of OLS
IV. Functional Form, Dummy Variables, and Index number
• Index number: GDP Example
• Base period: 1988
• Base value: 100
1988 1997 1998 1999
Nominal GDP 1,559.8 4,740.2 4,628.4 4,615.4 Real GDP 1,559.8 3,074.5 2743.4 2859.2 Index: GDP Deflator 100.0 154.2 168.7 161.4
I. Nature II. Examples III. Properties IV. Index 31
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Change the base year
• NESDB uses 1988 as the base year for GNP deflator 1988 100 (base year)
1997 154
1999 161
• Suppose we want to change the base year to 1997
1988 65 (100/154)*100
1997 100 (base year) (154/154)*100
1999 105 (161/154)*100
100 index *
index index old
base new
t
t
new old
I. Nature II. Examples III. Properties IV. Index 32
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Old and new GDP deflator Indexes compared.
1988 1997 1998 1999
Old GDP deflator Index 100.0 154.2 168.7 161.4
New GDP deflator Index 65 100 109 105
Calculate the growth rate of GDP deflator: inflation
100
* year )
year year
( new rate
growth
t t t
old
old
I. Nature II. Examples III. Properties IV. Index 33
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Calculate the growth rate of GDP deflator: another index for inflation
1988 1997 1998 1999
Old GDP deflator Index 100.0 154.2 168.7 161.4 New GDP deflator Index 64.9 100.0 109.4 104.7
Inflation rate 54.2 9.4 -4.3
Inflation in 1999 = [(104.7‐109.4)/109.4]*100
= ‐4.3%
I. Nature II. Examples III. Properties IV. Index 34
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Example : Effect of Personal Exemption on Fertility Rates;
gfr t = 0 + 1 pe t + 2 ww2 t + 3 pill t + u t
gfr t : general fertility rate – the number of children born to every 1,000 women.
pe t : real dollar value of the personal tax exemption ww2 = 1 during WW II (1941‐1945)
= 0 otherwise (1913‐40,1946‐84) pill = 1 from 1963 on (1963‐84)
= 0 otherwise (1913‐62)
Estimation:
t
= 98.68 + .083pe
t– 24.24ww2
t– 31.59pill
ts.e. (3.21) (.030) (7.46) (4.08)
t {30.74} {2.77} {3.25} {7.74}
n = 72 R
2= 0.473 R
2bar = 0.450 What can you say about statistical significance?
I. Nature II. Examples III. Properties IV. Index 35
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Dependent Variable: GFR Method: Least Squares Sample: 1 72
Included observations: 72
Variable Coefficient Std. Error t-Statistic Prob.
C 98.68176 3.208129 30.75991 0
PE 0.08254 0.029646 2.784166 0.0069
WW2 -24.2384 7.458253 -3.249876 0.0018
PILL -31.594 4.081068 -7.74161 0
R-squared 0.473415 Mean dependent var 95.63194
Adjusted R-squared 0.450184 S.D. dependent var 19.80464
S.E. of regression 14.68506 Akaike info criterion 8.265492
Sum squared resid 14664.27 Schwarz criterion 8.391973
Log likelihood -293.558 F-statistic 20.37801
Durbin-Watson stat 0.176873 Prob(F-statistic) 0
Fertility Rate Equation with no lags
I. Nature II. Examples III. Properties IV. Index 36
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Example: fertility rate equation with a lag
t
= 95.87 + .073pe
t‐ .0058pe
t‐1+ .034pe
t‐2‐ 22.31ww2
t‐ 31.30pill t‐stat {29.07} {.579} {‐.030} {.269} {11.58} {7.86}
n =70 R
2= 0.499 R
2‐bar= 0.459
• Interpretation:
1) Interpret the coefficient on ww2.
2) Interpret the coefficient on pill.
3) Are pe t , pe t‐1 , pe t‐2 individually statistically significant?
4) Are pe t , pe t‐1 , pe t‐2 jointly statistically significant?
– Problem: multicollinearity??
I. Nature II. Examples III. Properties IV. Index 37
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Dependent Variable: GFR Sample(adjusted): 3 72
Included observations: 70 after adjusting endpoints
Variable Coefficient Std. Error t-Statistic Prob.
C 95.8705 3.281957 29.21138 0
PE 0.072672 0.125533 0.578906 0.5647
PE(-1) -0.00578 0.155663 -0.037129 0.9705
PE(-2) 0.033827 0.126257 0.267919 0.7896
WW2 -22.1265 10.73197 -2.061737 0.0433
PILL -31.305 3.981559 -7.862495 0
R-squared 0.498599 Mean dependent var 94.77429
Adjusted R-squared 0.459427 S.D. dependent var 19.40881 S.E. of regression 14.27008 Akaike info criterion 8.236023 Sum squared resid 13032.64 Schwarz criterion 8.428751
Log likelihood -282.261 F-statistic 12.72845
Durbin-Watson stat 0.188715 Prob(F-statistic) 0
I. Nature II. Examples III. Properties IV. Index 38
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Wald Test:
Equation: Untitled
Test Statistic Value df Probability
F-statistic 0.05343 (2, 64) 0.948
Chi-square 0.10686 2 0.948
Null Hypothesis Summary:
Normalized Restriction (= 0) Value Std. Err.
C(3) -0.00578 0.155663
C(4) 0.033827 0.126257
Restrictions are linear in coefficients.
View/Coefficient Tests/Wald-Coefficient Restrictions
Redundant Variables: PE PE(-1) PE(-2)
F-statistic 3.972964 Probability 0.011652 Log likelihood ratio 11.95477 Probability 0.00754
View/Coefficient Tests/Redundant Variables…
I. Nature II. Examples III. Properties IV. Index 39
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Long run propensity...
gfr
t=
0+
0pe
t+
1pe
t‐1+
2pe
t‐2+
2ww2
t+
3pill
t+ u
tt
= 95.87 + .073pe
t‐ .0058pe
t‐1+ .034pe
t‐2‐ 22.31ww2
t‐ 31.30pill t {29.07} {.579} {‐.030} {.269} {11.58} {7.86}
5) Should we ditch pe t‐1 or pe t‐2 ?
6) What is the long‐run propensity in this model?
7) Is the long‐run propensity statistically significant?
What is the null hypothesis?
8) Find a 95% confidence interval of the long‐run propensity.
I. Nature II. Examples III. Properties IV. Index 40
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Dependent Variable: GFR Sample(adjusted): 3 72
Included observations: 70 after adjusting endpoints
Variable Coefficient Std. Error t-Statistic Prob.
C 95.8705 3.281957 29.21138 0
PE 0.100719 0.029803 3.379532 0.0012
PE(-1)-PE -0.00578 0.155663 -0.037129 0.9705
PE(-2)-PE 0.033827 0.126257 0.267919 0.7896
WW2 -22.1265 10.73197 -2.061737 0.0433
PILL -31.305 3.981559 -7.862495 0
R-squared 0.498599 Mean dependent var 94.77429
Adjusted R-squared 0.459427 S.D. dependent var 19.40881 S.E. of regression 14.27008 Akaike info criterion 8.236023 Sum squared resid 13032.64 Schwarz criterion 8.428751
Log likelihood -282.261 F-statistic 12.72845
Durbin-Watson stat 0.188715 Prob(F-statistic) 0
I. Nature II. Examples III. Properties IV. Index 41
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
The 95% CI of LRP
• The 95% confidence interval of LRP is
Given = 0.101 and s.e.( ) = 0.030, the 95% CI is
c*s.e.( ) = .101 2(.030)
= .041 to .160
• c: is the 95 th percentile in the t distribution with 64 DF n = 70 k = 5
n – k –1 = 70 –5 – 1= 64 c = 2.00 (DF = 60)
= 1.987 (DF = 90)
I. Nature II. Examples III. Properties IV. Index 42
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat IV. Functional Form, Dummy and Index
Recap of Time Series Analysis
• The Nature of Time Series Data
• Examples of Time Series Models
• Finite Sample Properties of OLS
• Functional Form, Dummy Variables, and Index Numbers
I. Nature II. Examples III. Properties IV. Index 43
11. Time Series Analysis . Quantitative Methods of Economic Analysis . 2949605 . Chairat Aemkulwat