ECON 2310 Stationary Time Series

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/112

encourage image

There's no tags or description

Looks like no tags are added yet.

113 Terms

1
New cards

Time series data

data collected on the same observational unit at multiple time periods

2
New cards

Examples of Time Series Data


3
New cards

logarithm of real gdp in US


4
New cards

growth rate of real gdp in us


5
New cards

why use time series data

forecasting models, dynamic causal effects, lack of choice

6
New cards

forecasting models

What will the rate of inflation be next year?

7
New cards

to estimate dynamic causal effects

If the Fed increases the Federal Funds rate now, what will be the effect on the rates of inflation and unemployment in 3 months? in 12 months?

– What is the effect over time on cigarette consumption of a hike in the cigarette tax

8
New cards

Plus, sometimes you don’t have any choice…

Rates of inflation and unemployment in the US can be observed only over time

9
New cards

learning about time series


10
New cards

time series data and technical issues

time lags

correlation over time

forecasting models w/o causal interpretation

conditions under which dynamic effects can be estimated

calculation of standard errors

11
New cards

correlation over time

serial correlation / autocorrelation

12
New cards

Forecasting models that have no causal interpretation (specialized tools for forecasting):

autoregressive (AR) models

– autoregressive distributed lag (ADL) models

13
New cards

Conditions under which dynamic effects can be estimated, and

how to estimate them

14
New cards

Calculation of standard errors when

the errors are serially correlated

15
New cards

Forecasting and estimation of causal effects are quite

different objectives

16
New cards

for forecasting


17
New cards

Time series data and serial correlation:

Notation and terminology on the time series 𝑌:


18
New cards

Time Series Data and Serial Correlation

We only consider which observations?

We only consider consecutive, evenly-spaced observations (e.g.: daily, weekly, monthly, quarterly, yearly; with no missing periods)

19
New cards

We will transform time series variables using:

Lags, First Differences, Logarithms, Growth Rates

20
New cards

The first and jth lag of a time series are


21
New cards

the first difference of a series


22
New cards

the first difference of the logarithm is


23
New cards

the percentage change of a time series is


24
New cards

Illustrating lags, changes, percentage changes using U.S. GDP data


25
New cards

2016 Q4 and 2017 Q1


26
New cards

4th column


27
New cards

last column

first lag, taking value from previous period

28
New cards

The correlation of a series with its own lagged values is

autocorrelation or serial correlation.

29
New cards

first autocorrelation of Yt


30
New cards

First autocovariance of Yt


31
New cards

autocorrelation and autocovariance link


32
New cards

jth autocovariance/autocorrelation description


33
New cards

jth autocovariance/autocorrelation formulas


34
New cards

the jth autocorrelation coefficient is sometimes called


35
New cards

sample autocovariance/autocorrelation


36
New cards

sample autocovariance/autocorrelation formula


37
New cards

sample autocovariance/autocorrelation anatomy


38
New cards

autocorrelation of the growth rate of US GDP example

1st four sample autocorrelations of GDP growth rate


39
New cards

notes on this


40
New cards

other examples


41
New cards

stationarity is the

key idea for external validity of time series regression

Historical relationships can be generalized to the future


42
New cards

stationarity definition


43
New cards

joint stationarity


44
New cards

stationarity requires the future to be

like the past, at least in a probabilistic sense

45
New cards

compute mean, variance, and jth autocovariance of Yt


46
New cards

predicted value


47
New cards

forecast


48
New cards

one step ahead forecast


49
New cards



50
New cards



51
New cards



52
New cards

forecast error


53
New cards



54
New cards

root mean squared forecast error


55
New cards

more on RMSFE


56
New cards

A natural starting point for a forecasting model is to use


57
New cards

autoregression


58
New cards

autoregression order


59
New cards

first order autoregressive model


60
New cards

AR(1) notes


61
New cards

AR(1) notes cont


62
New cards



63
New cards



64
New cards



65
New cards

AR(p) notes


66
New cards

AR(p) notes cont


67
New cards



68
New cards

Autoregressive Distributed Lag (ADL) Model


69
New cards

notes on ADL model


70
New cards

Example: GDP growth rate and term spread


71
New cards

Term Spread:

difference between long-term and short-term interest rates

72
New cards

Short-term rates are generally

below long-term rates, but the gap narrows shortly before the start of a recession

73
New cards



74
New cards



75
New cards

coefficient on TSpreadt-1 and adjusted R2


76
New cards

exercise: compute forecast error


77
New cards



78
New cards

gdp growth rate and term spread coefficient on TSpreadt-2 and adjusted R2


79
New cards

f we know GDPGR in 2017:Q3 = 3.11, GDPGR in 2017


80
New cards

equations 15.14 and 15.15

Eq. (15.14) is an ADL(2,1) model, Eq. (15.15) is an ADL(2,2) model

81
New cards

general time series regression model definition


82
New cards

general time series regression model math


83
New cards

general time series regression model anatomy


84
New cards

Least Squares Assumptions for Forecasting with Multiple Predictors

They are time series counterparts of the four least squares assumptions for prediction with multiple regression using cross-sectional data

85
New cards

assumption 2 general time series regression

identically and independently distrubuted

86
New cards

identically distributed


87
New cards

independently distributed


88
New cards

estimate of MSFE can be used


89
New cards

MSFE AR(p) model


90
New cards



91
New cards



92
New cards

method 2 Substitution of this expression into MSFE gives us:


93
New cards



94
New cards

method 3 first 2 steps


95
New cards

method 3 second part


96
New cards

anatomy


97
New cards



98
New cards

advantage/disadvantage of method 3


99
New cards

Example: Pseudo Out-Of-Sample MSFE estimator

step 1


100
New cards

Example: Pseudo Out-Of-Sample MSFE estimator

step 2-4