An Introduction to ECMs

An Introduction to Error Correction Models

Robin Best Oxford Spring School for Quantitative Methods in Social Research 2008

Error Correction Models (ECMs) are a category of multiple time series models that directly estimate the speed at which a dependent variable Y - returns to equilibrium after a change in an independent variable - X. ECMs are useful for estimating both short term and long term effects of one time series on another.
Thus, they often mesh well with our theories of political and social processes. Theoretically-driven approach to estimating time series models.

ECMs are useful models when dealing with integrated data, but can also be used with stationary data.

An Introduction to ECMs
The basic structure of an ECM Yt = + Xt-1 - ECt-1 + t
Where EC is the error correction component of the model and measures the speed at which prior deviations from equilibrium are corrected.

An Introduction to ECMs
As we will see, the versatility of ECMs give them a number of desirable properties.
Estimates of short and long term effects Easy interpretation of short and long term effects Applications to both integrated and stationary time series data

Error correction models can be used to estimate the following quantities of interest for all X variables.
Short term effects of X on Y Long term effects of X on Y (long run multiplier) The speed at which Y returns to equilibrium after a deviation has occurred.

Can be estimated with OLS Model theoretical relationships

ECMs can be appropriate whenever (1) we have time series data and (2) are interested in both short and long term relationships between multiple time series.

Applications of ECMs in the (Political Science) Literature

U.S. Presidential Approval/ U.K. Prime Ministerial Satisfaction Policy Mood/Policy Sentiment Support for Social Security Consumer Confidence Economic Expectations Health Care Cost Containment/ Government Spending /Patronage Spending / Redistribution Interest Rates/ Purchasing Power Parity Growth in (U.S.) Presidential Staff Arms Transfers U.S. Judicial Influence

Overview of the Course

I.

Motivating ECMs with cointegrated data

Integration and cointegration 2-step error correction estimators Stata session #1

II.

Motivating ECMs with stationary data

The single equation ECM Interpretation of long and short term effects The Autoregressive Distributive Lag (ADL) model Equivalence of the ECM and ADL Stata session #2

ECMs and Cointegration: Stationary vs. Integrated Time Series

Stationary time series data are mean reverting. That is, they have a finite mean and variance that do not depend on time.

ECMs and Cointegration: Stationary vs. Integrated Time Series

Often our time series data are not stationary, but appear to be integrated. Integrated time series data
Are not mean-reverting appear to be on a random walk Have current values that can be expressed as the sum of all previous changes The effect of any shock is permanently incorporated into the series Thus, the best predictor of the series at time t is the value at time t-1 Have a (theoretically) infinite variance and no mean.

Yt = + Yt-1 + t
Where | p | < 1 and t is also stationary with a mean of zero and variance 2

Note that when 0 < | p | < 1 the time series is stationary but contains autocorrelation.

ECMs and Cointegration: Integrated Time Series

Formally, an integrated series can be expressed as a function of all past disturbances at any point in time.
t

ECMs and Cointegration: Integrated Time Series

Order of Integration
Integrated time series data that are stationary after being difference d times are Integrated of order d: I(d)

Yt

ei
i =1

Or Yt = + Yt-1 + t Where p = 1

For our purposes, we focus on time series data that are I(1). Data that are stationary after being first-differenced. I(1) processes are fairly common in time series data

Or Yt - Yt-1 = ut Where ut = t And t is still a stationary process

ECMs and Cointegration: Integrated Time Series

(Theoretical) Sources of integration

A Drunks Random Walk

The effect of past shocks is permanently incorporated into the memory of the series. The series is a function of other integrated processes.

20 time

40

60

ECMs and Cointegration: Integrated Time Series

ECMs and Cointegration: Integrated Time Series

Analyzing time series data in differenced form solves the spurious
regression problem, but may throw the baby out with the bathwater.

Analyzing integrated time series in level form dramatically increases the likelihood of making a Type-II error. Problem of spurious associations. High R2

A model that includes only (lagged) differenced variables assumes the

effects of the X variables on Y never last longer than one time period.

Small standard errors and inflated t-ratios

What if our time series share a long run relationship? If the time series share an equilibrium relationship with an errorcorrection mechanism, then the stochastic trends of the time series will be correlated with one another. Cointegration

A common solution to these problems is to analyze the data in differenced form. Look only at short term effects

ECMs and Cointegration

Two time series are cointegrated if Both are integrated of the same order.

ECMs and Cointegration

Lets go back to the drunks random walk and call the drunk X. The random walk can be expressed as Xt - Xt-1 = ut

There is a linear combination of the two time series that is I(0) - i.e. stationary. Two (or more) series are cointegrated if each has a long run component, but these components cancel out between the series.
Share stochastic trends

Where ut represents the stationary, white-noise shocks.

Another rather trivial example of a random walk is the walk (or jaunt) of a dog, which can be expressed as Yt - Yt-1 = wt Where wt represents the stationary, while-noise process of the dogs steps.

Conintegrated data are never expected to drift too far away from each other, maintaining an equilibrium relationship.

A Dogs Random Walk

ECMs and Cointegration

But what if the dog belongs to the drunk?
Then the two random walks are likely to have an equilibrium relationship and to be cointegrated (Murray 1994). Deviations from this equilibrium relationship will be corrected over time. Thus, part of the stochastic processes of both walks will be shared and will correct deviations the equilibrium

Xt - Xt-1 = ut + c(Yt-1 - Xt-1) Yt - Yt-1 = wt + d(Xt-1 - Yt-1)

0 20 time 40 60

Where the terms in parentheses are the error correcting mechanisms

The Drunk and Her Dog

ECMs and Cointegration

Two I(1) time series (Xt and Yt) are cointegrated if there is some linear combination that is stationary. Zt = Yt - Xt
Where Z is the portion of (levels of) Y that are not shared with X: the equilibrium errors. We can also rewrite this equation in regression form

Yt = Xt + Zt
0 20 time drunk dog 40 60

Where the cointegrating vector - Zt - can be obtained by regressing Yt on Xt.

ECMs and Cointegration

Yt = Xt + Zt
Here, Z represents the portion of Y (in levels) that is not attributable to X. In short, Z will capture the error correction relationship by capturing the degree to which Y and X are out of equilibrium.

ECMs and Cointegration

Yt will be a function of the degree to which the two time series were out of equilibrium in the previous period: Zt-1 Zt-1 = Yt-1 - Xt-1
When Z = 0 the system is in its equilibrium state Yt will respond negatively to Zt-1.

Z will capture any shock to either Y or X. If Y and X are cointegrated, then the relationship between the two will adjust accordingly.

If Z is negative, then Y is too high and will be adjusted downward in the next period. If Z is positive, then Y is too low and will be adjusted upward in the next time period.

ECMs and Cointegration

We might theorize that shocks to X have two effects on Y.
Some portion of shocks to X might immediately affect Y in the next time period, so that Yt responds to Xt-1. A shock to Xt will also disturb the equilibrium between Y and X, sending Y on a long term movement to a value that reproduces the equilibrium state given the new value of X. Thus Yt is a function of both Xt-1 and the degree to which the two variables were out of equilibrium in the previous time period.

Engle and Granger Two-Step ECM

If two time series are integrated of the same order AND some linear combination of them is stationary, then the two series are cointegrated. Cointegrated series share a stochastic component and a long term equilibrium relationship. Deviations from this equilibrium relationship as a result of shocks will be corrected over time. We can think of Yt as responding to shocks to X over the short and long term.

Engle and Granger Two-Step ECM

Step 1: Engle and Granger (1987) suggested an appropriate model for Y, based two or more time series that are cointegrated.

Engle and Granger Two-Step ECM

Yt = + Xt + Zt
The cointegrating vector - Z - is measured by taking the residuals from the regression of Yt on Xt

First, we can obtain an estimate of Z by regressing Y on X. Step 2: Second, we can regress Yt on Zt-1 plus any relevant short term effects.

Zt = Yt - Xt -

Regress changes on Y on lagged changes in X as well as the equilibrium errors represented by Z.

Yt = 0 Xt-1 - 1Zt-1
Note that all variables in this model are stationary.

Engle and Granger Two-Step ECM

In Step 1, where we estimate the cointegrating regression we can and should - include all variables we expect to 1) be cointegrated 2) have sustained shocks on the equilibrium. The variables that have sustained shocks on the equilibrium are usually regarded as exogenous shocks and often take the form of dummy variables. Then

Engle and Granger Two-Step ECM

The cointegrating regression is performed as Yt = + Xt + Zt Which we can also conceptualize as Zt = Yt - ( +Xt) If we add a series of j exogenous shocks - represented as wj Yt = + Xt+ W 1t + W 2t +W 3t + Zt Zt = Yt - ( +Xt + W 1t + W 2t +W 3t)

Engle and Granger Two-Step ECM

The basic structure of the ECM

Engle and Granger Two-Step ECM

Note that the Engle and Granger 2-Step method is really a 4-step method. 1) Determine that all time series are integrated of the same order. 2) Demonstrate that the time series are cointegrated

Yt = + Xt-1 - ECt-1 + t
In the Engle and Granger Two-Step Method the EC component is derived from cointegrated time series as Z.

Yt = 0 Xt-1 - 1Zt-1
0 captures the short term effects of X in the prior period on Y in the current period. 1 captures the rate at which the system Y adjusts to the equilibrium state after a shock. In other words, it captures the speed of error correction.

3) Obtain an estimate of the cointegrating vector - Z - by regressing Yt on Xt and taking the residuals. 4) Enter the lagged residuals - Z - into a regression of Yt on Xt-1.

Engle and Granger Two-Step ECM

Viewed from this perspective, it is easy to see why error correction models have become so closely associated with cointegration (we will come back to this later). Integrated time series present a problem for time series analysis - at least in terms of long term relationships. When integrated time series variables are also cointegrated, error correction models provide a nice solution to this problem.

Cointegration and Error Correction

One of the first instances of error correction was Davidson et. al.s (1978) study of consumer expenditure and income in the U.K.. The Engle and Granger approach to error correction models follows nicely from the field of economics, where integration and cointegration among time series is theoretically common. Error correction models were imported from economics. Would we expect data from the social sciences to follow similar patterns of integration and cointegration?

Cointegration and Error Correction in Political Science

Prime Ministerial Statisfaction (U.K.) and Conservative Party Support Arms transfers by the U.S. and Soviet Union Economic expectations and U.S. Presidential Approval U.S. Domestic Policy Sentiment and Economic Expectations Support for U.S. Social Security and the Stock Market

The Engle and Granger Two-Step ECM: Putting it into Practice

Lets imagine we have two time series - perhaps the drunk and her dog but lets call the drunk X and the dog Y. From a theoretical perspective, we believe changes in X will have both short and long term effects on Y, since we expect X and Y to have an equilibrium relationship.

We expect changes in X to produce long run responses in Y, as Y adjusts back to the equilibrium state.

X and Y: Cointegrated?
25 20

Engle and Granger Two-Step ECM

First, we need to determine that both X and Y are integrated of the same order. Which means we first need to demonstrate that both X and Y are, in fact, integrated processes. We should also think about the likely stationary or nonstationary nature of our time series from a theoretical perspective.

10

15

Tests for unit-root process tend to be controversial, primarily due to their low power.
5

For our purposes, we will focus on Dickey-Fuller (DF) and Augmented Dickey-Fuller tests to examine the (non)stationarity of our time series.
1961m1 1962m1 1963m1 months Y X 1964m1 1965m1 0 1960m1

Dickey-Fuller Tests
Basic Dickey-Fuller test

Dickey-Fuller Tests
Basic Dickey-Fuller test With a constant (drift) With a time trend

xt = xt 1 + t

xt = xt 1 + t

xt = t + xt 1 + t

With a constant (drift)

xt = t + xt 1 + t
xt = t + xt 1 + t + t

xt = t + xt 1 + t + t
If X is a random walk process, then = 0 The null hypothesis is that X is a random walk MacKinnon values for statistical significance

With a time trend

Note that in small samples the standard error of will be large, making it likely that we fail to reject the null when we really should

Augmented Dickey-Fuller
We can remove any remaining serial correlation in t by introducing an appropriate number of lagged differences of X in the equation.
k

Is X Integrated?
dfuller X, regress Dickey-Fuller test for unit root Number of obs = 63

xt = xt 1 + i x1t i + t
i =1

xt = t + xt 1 + i x1t i + t
i =1

---------- Interpolated Dickey-Fuller --------Test 1% Critical 5% Critical 10% Critical Statistic Value Value Value -----------------------------------------------------------------------------Z(t) -1.852 -3.562 -2.920 -2.595 -----------------------------------------------------------------------------MacKinnon approximate p-value for Z(t) = 0.3548 -----------------------------------------------------------------------------D.X | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------X | L1. | -.1492285 .0805656 -1.85 0.069 -.3103293 .0118724 _cons | 1.365817 .7149307 1.91 0.061 -.0637749 2.79541 ------------------------------------------------------------------------------------------------------------------------------------------------

Where i = 1, 2, k Null hypotheses are the same as the DF tests

Is X Integrated?
dfuller X, lags(4) regress Number of obs = 59 Augmented Dickey-Fuller test for unit root ---------- Interpolated Dickey-Fuller --------Test 1% Critical 5% Critical 10% Critical Statistic Value Value Value -----------------------------------------------------------------------------Z(t) 0.690 -3.567 -2.923 -2.596 -----------------------------------------------------------------------------MacKinnon approximate p-value for Z(t) = 0.9896 -----------------------------------------------------------------------------D.X | Coef. Std. Err. t P>|t| [95% Conf.Interval] -------------+---------------------------------------------------------------X | L1. | .0696672 .1008978 0.69 0.493 -.1327082 .2720426 LD. | -.5724812 .1738494 -3.29 0.002 -.9211789 .2237835 L2D. | -.4935811 .1776346 -2.78 0.008 -.8498709 -.1372912 L3D. | -.2891465 .1677748 -1.72 0.091 -.6256601 .0473671 L4D. | -.0898266 .1468121 -0.61 0.543 -.3842943 .2046412 _cons | -.2525666 .839646 -0.30 0.765 -1.936683 1.43155 ------------------------------------------------------------------------------

Is X Integrated?
If X is I(1), then the first difference of X should be stationary.

dfuller dif_X Dickey-Fuller test for unit root Number of obs = 62

---------- Interpolated Dickey-Fuller --------Test 1% Critical 5% Critical 10% Critical Statistic Value Value Value -----------------------------------------------------------------------------Z(t) -10.779 -3.563 -2.920 -2.595 -----------------------------------------------------------------------------MacKinnon approximate p-value for Z(t) = 0.0000

Is Y Integrated?
dfuller Y, regress Dickey-Fuller test for unit root Number of obs = 63

Is Y Integrated?
dfuller dif_Y, regress Dickey-Fuller test for unit root Number of obs = 62 ---------- Interpolated Dickey-Fuller --------Test 1% Critical 5% Critical 10% Critical Statistic Value Value Value -----------------------------------------------------------------------------Z(t) -9.071 -3.563 -2.920 -2.595 -----------------------------------------------------------------------------MacKinnon approximate p-value for Z(t) = 0.0000 -----------------------------------------------------------------------------D.dif_Y | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------dif_Y | L1. | -1.159903 .1278662 -9.07 0.000 -1.415674 -.9041329 _cons | .2219184 .3259962 0.68 0.499 -.4301711 .8740078 ------------------------------------------------------------------------------

---------- Interpolated Dickey-Fuller --------Test 1% Critical 5% Critical 10% Critical Statistic Value Value Value -----------------------------------------------------------------------------Z(t) -1.323 -3.562 -2.920 -2.595 -----------------------------------------------------------------------------MacKinnon approximate p-value for Z(t) = 0.6184 -----------------------------------------------------------------------------D.Y | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------Y | L1. | -.0854922 .064599 -1.32 0.191 -.2146659 .0436814 _cons | 1.061271 .7208156 1.47 0.146 -.3800884 2.502631 ------------------------------------------------------------------------------

Cointegration
Both X and Y appear to be integrated of the same order: I(1). If they are cointegrated, then they share stochastic trends. In the following regression, t should be stationary and should be statistically significant and in the expected direction.

Cointegrating Regression
regress Y X Source | SS df MS -------------+-----------------------------Model | 1009.22604 1 1009.22604 Residual | 676.523964 62 10.9116768 -------------+-----------------------------Total | 1685.75 63 26.7579365 Number of obs F( 1, 62) Prob > F R-squared Adj R-squared Root MSE = = = = = = 64 92.49 0.0000 0.5987 0.5922 3.3033

Yt = t + Xt +t
Lets see if this is the case

-----------------------------------------------------------------------------Y | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------X | 1.206126 .1254135 9.62 0.000 .9554281 1.456824 _cons | .0108108 1.135884 0.01 0.992 -2.259789 2.28141 ------------------------------------------------------------------------------

Cointegrating Regression
predict r, resid

dfuller r Dickey-Fuller test for unit root Number of obs = 63

---------- Interpolated Dickey-Fuller --------Test 1% Critical 5% Critical 10% Critical Statistic Value Value Value -----------------------------------------------------------------------------Z(t) -5.487 -3.562 -2.920 -2.595 -----------------------------------------------------------------------------MacKinnon approximate p-value for Z(t) = 0.0000

-15 1960m1

-10

Residuals -5 0

10

1961m1

1962m1

1963m1 months

1964m1

1965m1

Engle and Granger Two-Step ECM

Our residuals from the cointegrating regression capture deviations from the equilibrium of X and Y. Therefore, we can estimate both the short and long term effects of X on Y by including the lagged residuals from the cointegrating regression as our measure of the error correction mechanism.

Engle and Granger Two-Step ECM

regress dif_Y dlag_X lag_r Source | SS df MS -------------+-----------------------------Model | 59.4494524 2 29.7247262 Residual | 344.227967 59 5.83437232 -------------+-----------------------------Total | 403.677419 61 6.61766261 Number of obs = 62 F( 2, 59) = 5.09 Prob > F = 0.0091 R-squared = 0.1473 Adj R-squared = 0.1184 Root MSE = 2.4154

Yt = + 1*Xt-1 + 2*Rt-1 +t

-----------------------------------------------------------------------------dif_Y | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------dlag_X | -.1161038 .1609359 -0.72 0.473 -.4381358 .2059282 lag_r | -.3160139 .0999927 -3.16 0.002 -.5160988 -.1159291 _cons | .210471 .3074794 0.68 0.496 -.4047939 .8257358 ------------------------------------------------------------------------------

The error correction mechanism is negative and significant, suggesting that deviations from equilibrium are corrected at about 32% per month. However, X does not appear to have significant short term effects on Y.

Granger Causality and ECMs

Granger Causality: A variable - X Granger causes another variable Y if Y can be better predicted by the lagged values of both X and Y than by the lagged values of Y alone (see Freeman 1983). Standard Granger causality tests can result in incorrect inferences about causality when there is an error correction process. The Engle-Granger approach to ECMs begins by assuming all variables in the cointegrating regression are jointly endogeneous. Thus, in the previous example we should also estimate a cointegrating regression of X on Y.

Granger Causality
Granger causality can be ascertained in the ECM framework by
regressing each time series in differenced form on all time series in both differenced and level form.

If an EC representation is appropriate, then in at least one of the

regressions: The lagged level of the predicted variable should be negative and significant. The lagged level of the other variable should be in the expected direction and significant.

Granger Causality
regress dif_Y l.dif_Y l.dif_X lag_Y lag_X Source | SS df MS -------------+-----------------------------Model | 69.5277246 4 17.3819311 Residual | 334.149695 57 5.86227535 -------------+-----------------------------Total | 403.677419 61 6.61766261 Number of obs = 62 F( 4, 57) = 2.97 Prob > F = 0.0270 R-squared = 0.1722 Adj R-squared = 0.1141 Root MSE = 2.4212

Granger Causality
regress dif_X l.dif_X l.dif_Y lag_X lag_Y Number of obs = F( 4, 57) Prob > F R-squared Adj R-squared Root MSE 62 5.87 0.0005 0.2917 0.2420 1.7779 Source | SS df MS -------------+-----------------------------Model | 74.2042429 4 18.5510607 Residual | 180.182854 57 3.1611027 -------------+-----------------------------Total | 254.387097 61 4.17028027 = = = = =

-----------------------------------------------------------------------------dif_Y | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------dif_Y | L1. | .0483244 .1399056 0.35 0.731 -.2318318 .3284806 dif_X | L1. | -.2205689 .1802099 -1.22 0.226 -.581433 .1402952 lag_Y | -.3557259 .1161894 -3.06 0.003 -.5883911 -.1230606 lag_X | .5675793 .1899981 2.99 0.004 .1871146 .948044 _cons | -.928984 .9426534 -0.99 0.329 -2.816615 .9586468 ------------------------------------------------------------------------------

-----------------------------------------------------------------------------dif_X | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------dif_X | L1. | -.0640245 .132332 -0.48 0.630 -.3290147 .2009657 dif_Y | L1. | .0014809 .1027357 0.01 0.989 -.2042438 .2072056 lag_X | -.4676537 .1395197 -3.35 0.001 -.7470371 -.1882703 lag_Y | .2847586 .0853204 3.34 0.001 .1139075 .4556097 _cons | 1.194109 .6922106 1.73 0.090 -.1920183 2.580237 ------------------------------------------------------------------------------

ECMs, Causality, and Theory

In the social sciences, our theories (usually) tell us which time series should be on the left side of the equation and which should be on the right.

Engle and Granger Two-Step Technique: Issues and Limitations

Does not clearly distinguish dependent variables from independent variables.
In the social sciences the Engle and Granger two-step ECM might not be consistent with our theories.

The Engle and Granger approach assumes endogeneity between the cointegrating time series.

Is appropriate when dealing with cointegrated time series.

Can we clearly distinguish between integrated and stationary processes?

Integration Issues
Error correction approaches that rely on cointegration of two or more I(1) time series become problematic when we are dealing with data that are not truly (co)integrated. I(1) processes may be incorrectly included into the cointegrating regression - producing spurious associations - if two other I(1) cointegrated time series are already included (Durr 1992)
This problem increases with sample size.

More Integration Issues

In the social sciences, we are more likely to have data that are
Near integrated (p = 0, but there is memory. p may not = 0 in finite samples.) Fractionally integrated (0 < p < 1, where when 0 < p < .5 the data are mean-reverting and have finite variance, and when .5 p < 1 the data are mean-reverting but have infinite variance) A combined process of both stationary and integrated data
Aggregated data

The low power of unit root tests can lead us to conclude our data are integrated when they are not.

More Integration Issues

Under these conditions, we are likely to draw faulty inferences from the two-step procedure. We might conclude:

Integration Issues and ECMs

Under these conditions, we are often better off estimating a single equation ECM. Single equation ECMs solve some of these problems and avoid others.

Our data are integrated when they are not. Our data are cointegrated when they are not. Our data are not cointegrated, therefore, an ECM is not appropriate

However, single equation ECMs require weak exogeneity.

10

Single Equation Error Correction Models

Following theory, Single Equation ECMs clearly distinguish between dependent and independent variables.

Single Equation ECMs

Our theories might specify long and short term effects of independent variables on a dependent variable even when our data are stationary. The concepts of error correction, equilibrium , and long term effects are not unique to cointegrated data. Furthermore, an ECM may provide a more useful modeling technique for stationary data than alternative approaches. Our theories may be better represented by a single equation ECM.

Single Equation ECMs are appropriate for both cointegrated and longmemoried, but stationary, data.
Cointegration may imply error correction, but does error correction imply cointegration?

Single Equation ECMs estimate a long term effect for each independent variable, allowing us to judge the contribution of each.
Allow for easier interpretation of the effects of the independent variables.

Single Equation ECMs

Single Equation Error Correction Models are useful
When our theories dictate the causal relationships of interest When we have long-memoried/stationary data

The Single Equation ECM

Basic form of the ECM Yt = + Xt-1 - ECt-1 + t Engle and Granger two-step ECM Yt = 0 Xt-1 - 1Zt-1

A basic single equation ECM:

Where Zt = Yt - Xt -

Yt = + 0* Xt - 1(Yt-1 - 2Xt-1) + t The Single Equation ECM Yt = + 0* Xt - 1(Yt-1 - 2Xt-1) + t

The Single Equation ECM

Yt = + 0*Xt - 1(Yt-1 - 2Xt-1) + t
The portion of the equation in parentheses is the error correction mechanism. (Yt-1 - 2Xt-1) = 0 when Y and X are in their equilibrium state
0 estimates the short term effect of an increase in X on Y

The Single Equation ECM

Yt = + 0*Xt - 1(Yt-1 - 2Xt-1) + t
The values for which Y and X are in their long term equilibrium relationship are Y = k0 + k1X Where k 0 = 1 And

1 estimates the speed of return to equilibrium after a deviation.

k1 =

2 1

If the ECM approach is appropriate, then -1 < 1 < 0

2 estimates the long term effect that a one unit increase in X has on Y. This long term effect will be distributed over future time periods according to the rate of error correction - 1

Where k1 is the total long term effect of X on Y (a.k.a the long run multiplier) - distributed over future time periods. Single equation ECMs are particularly useful for allowing us to also estimate k1s standard error, and therefore statistical significance.

11

The Single Equation ECM

Since the long term effect is a ratio of two coefficients, we could calculate its standard error using the variance and covariance matrix Alternatively, we can use the Bewley transformation to estimate the standard error. This requires estimating the following regression.

The Single Equation ECM

We can easily extend the single equation ECM to include more independent variables

Yt = + X1t + X2t + X3t - (Yt-1 - X1t-1 - X2t-1 - X3t-1) + t

Yt = + 0 Yt + 1Xt - 2Xt + t
Where 1 is the long term effect and is estimated with a standard error Notice the problem: we have Yt on the right side of the equation We can proxy Yt as:

Note that each independent variable is now forced to make an independent contribution to the long term relationship, solving one of the problems in the two-step estimator.

Yt = + Yt-1 + Xt + Xt + t

And use our predicted values of Yt in the Bewley transformation regression

Single Equation ECMs in the (Political Science) Literature

Judicial Influence Health Care Cost Containment Interest Rates Patronage Spending Growth in Presidential Staff Government Spending Consumer Confidence Redistribution

Single Equation ECMs

Single Equation ECMs
Provide the same information about the rate of error correction as the Engle and Granger two-step method. Provide more information about the long term effect of each independent variable - including its standard error - than the Engle and Granger twostep method. Illustrate that ECMs are appropriate for both cointegrated and stationary data.

How do we know Single Equation ECMs are appropriate with stationary data?

ECMs and ADL Models

We know Autoregressive Distributive Lag models are appropriate for stationary data (stationary data is, in fact, a requirement of these models). Forms of single equation ECMs and ADL models are equivalent. We can derive a single equation ECM from a general ADL model:

ECMs and the ADL

Yt = + 0Yt-1 + 1Xt + 2Xt-1 + t Yt = + (0 - 1)Yt-1+ 1Xt + 2Xt-1 + t Yt = + (0 - 1)Yt-1+ 1Xt + (1 + 2)Xt-1 + t

Yt = + 0Yt-1 + 1Xt + 1Xt-1 + t

Yt = + 0Yt-1 + 1Xt + 2Xt-1 + t

Where 0 = 0 - 1 and 1 = 1 + 2 We can rewrite this equation in error correction form as

Yt = + 1Xt - 0(Yt-1 - 1Xt-1) + t

12

ECMs and the ADL

We can see that the ADL model provides information similar to the ECM.

ECMs and the ADL

Yt = + 0Yt-1 + 1Xt + 2Xt-1 + t
And the total long term effect/long run multiplier - k1 - is therefore:

Yt = + 0Yt-1 + 1Xt + 2Xt-1 + t

0 estimates the proportion of the deviation from equilibrium at t-1 that is maintained at time t. 0 - 1 tells us the speed of return. 1 estimates the short term effect of X on Y

k1 =

2 + 1 1 0

Y and X will be in their long term equilibrium state when Y = k0 + k1X where
k0 =

1 0

1 + 2 estimates the long term effect of a unit change in X on Y (the coefficient on Xt-1 in the ECM)

ECMs and ADL Models

What does this mean? ECMs are isophormic to ADL models
We can use them with stationary data

The EC and ADL Models: Notation

Lets use the following notation for the single equation ECM and the ADL

ECM

Yt = + 0Xt - 1(Yt-1 - 2Xt-1) + t

Certain forms of ADL models are - in a general sense - error correction models. They can be used to estimate:
The speed of return to equilibrium after a deviation has occurred. Long term equilibrium relationships between variables. Long and short term effects of independent variables on the dependent variable.

ADL

Yt = + 0Yt-1 + 1Xt + 2Xt-1 + t

Single Equation ECM

Lets imagine our theory about the relationship between X and Y states: X causes Y. X should have both a short term and a long term effect on Y. We dont have reason to suspect cointegration from a theoretical standpoint. But we believe X and Y share a long term equilibrium relationship

Single Equation ECM

We determine that our Y variable is stationary (with 95% confidence), ruling out an ECM based on cointegration
dfuller y, regress Dickey-Fuller test for unit root Number of obs = 55

---------- Interpolated Dickey-Fuller --------Test 1% Critical 5% Critical 10% Critical Statistic Value Value Value -----------------------------------------------------------------------------Z(t) -3.353 -3.573 -2.926 -2.598 -----------------------------------------------------------------------------MacKinnon approximate p-value for Z(t) = 0.0127

13

Single Equation ECM

We then estimate the single equation ECM

Single Equation ECM

regress dif_y dif_x lag_y lag_x Source | SS df MS -------------+-----------------------------Model | 238.216589 3 79.4055296 Residual | 189.278033 51 3.71133398 -------------+-----------------------------Total | 427.494622 54 7.91656707 Number of obs = 55 F( 3, 51) = 21.40 Prob > F = 0.0000 R-squared = 0.5572 Adj R-squared = 0.5312 Root MSE = 1.9265

Yt = + 0Xt - 1(Yt-1 - 2Xt-1) + t

As

Yt = + 0Xt + 1Yt-1 + 2Xt-1 + t

If our error correction approach is correct, then 1 should be -1 < 1 < 0 and significant.

-----------------------------------------------------------------------------dif_y | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------dif_x | 1.324821 .200003 6.62 0.000 .9232986 1.726344 lag_y | -.4248235 .1146587 -3.71 0.001 -.6550105 -.1946365 lag_x | .5182186 .1971867 2.63 0.011 .1223498 .9140873 _cons | 13.12112 4.201755 3.12 0.003 4.685745 21.55649 ------------------------------------------------------------------------------

Single Equation ECM

The results indicate the following equation

Single Equation ECM

Yt = 13.12 + 1.32*Xt -.42(Yt-1 - 1.22*Xt-1) + t
Y and X are in their long term equilibrium state when Y = 30.89 + 1.22X So that when X = 1 Y = 32.11

Yt = 13.12 + 1.32* Xt -.42*Yt-1 + .52*Xt-1 + t

Which we can write in error correction form as

Yt = 13.12 + 1.32* Xt -.42(Yt-1 - 1.22*Xt-1) + t

Where 1.22 is our calculation of the long run multiplier

Single Equation ECM

Yt = + 1.32* Xt -.42(Yt-1 - 1.22*Xt-1) + t
Changes in X have both an immediate and long term effect on Y When the portion of the equation in parentheses = 0, X and Y are in their equilibrium state. Increases in X will cause deviations from this equilibrium, causing Y to be too low. Y will then increase to correct this disequilibrium, with 42% of the (remaining) deviation corrected in each subsequent time period.

Single Equation ECM

Yt = + 1.32* Xt -.42(Yt-1 - 1.22*Xt-1) + t
A one unit increase in X immediately produces a 1.32 unit increase in Y. Increases in X also disrupt the the long term equilibrium relationship between these two variables, causing Y to be too low. Y will respond by increasing a total of 1.22 points, spread over future time periods at a rate of 42% per time period.
Y will increase .52 points at t Then another .3 points at t+1 Then another .2 points at t+2 Then another .1 points at t+3 Then another .05 points at t+4 Then another .03 points at t+5 Until the change in X at t-1 has virtually no effect on Y

14

1.5

Change in Y

Y
0 0 2 Time Period 4 6

1 0

1.5

.5

2.5

Time Period

Single Equation ECM

We can determine the standard error and confidence level of the total long term effect of X on Y through the Bewley transformation regression. First, we can obtain an estimate of Y by estimating Yt = + Yt-1 + Xt + Xt + t
regress dif_y lag_y x dif_x Source | SS df MS -------------+-----------------------------Model | 238.216589 3 79.4055296 Residual | 189.278033 51 3.71133398 -------------+-----------------------------Total | 427.494622 54 7.91656707 Number of obs F( 3, 51) Prob > F R-squared Adj R-squared Root MSE = = = = = = 55 21.40 0.0000 0.5572 0.5312 1.9265

Single Equation ECM

And take the predicted values of Yt to estimate Yt = + 0Yt + 1Xt - 2Xt + t
predict deltaYhat regress y deltaYhat x dif_x Source | SS df MS -------------+-----------------------------Model | 531.551099 3 177.1837 Residual | 189.278039 51 3.7113341 -------------+-----------------------------Total | 720.829138 54 13.3486877 Number of obs F( 3, 51) Prob > F R-squared Adj R-squared Root MSE = = = = = = 55 47.74 0.0000 0.7374 0.7220 1.9265

-----------------------------------------------------------------------------dif_y | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------lag_y | -.4248235 .1146587 -3.71 0.001 -.6550105 -.1946365 x | .5182186 .1971867 2.63 0.011 .1223498 .9140873 dif_x | .8066027 .2278972 3.54 0.001 .34908 1.264125 _cons | 13.12112 4.201755 3.12 0.003 4.685745 21.55649 ------------------------------------------------------------------------------

-----------------------------------------------------------------------------y | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------deltaYhat | -1.353919 .2698973 -5.02 0.000 -1.89576 -.8120773 x | 1.219844 .1245296 9.80 0.000 .9698408 1.469848 dif_x | 1.898677 .3963791 4.79 0.000 1.102913 2.694442 _cons | 30.88605 2.68463 11.50 0.000 25.49643 36.27567 ------------------------------------------------------------------------------

Single Equation ECM

We can see our estimate of the long term effect of X on Y has a standard error of .12 and is statistically significant.

Equivalence of the EC and ADL models

First, lets estimate Yt = + 0Yt-1 + 1Xt + 2Xt-1 + t
regress y lag_y x lag_x Source | SS df MS -------------+-----------------------------Model | 531.551105 3 177.183702 Residual | 189.278033 51 3.71133398 -------------+-----------------------------Total | 720.829138 54 13.3486877 Number of obs = F( 3, 51) Prob > F R-squared Adj R-squared Root MSE 55 47.74 0.0000 0.7374 0.7220 1.9265

= = = = =

Can we gain similar estimates of the short and long term effects of X on Y from the ADL model?
-----------------------------------------------------------------------------y | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------lag_y | .5751765 .1146587 5.02 0.000 .3449895 .8053635 x | 1.324821 .200003 6.62 0.000 .9232986 1.726344 lag_x | -.8066027 .2278972 -3.54 0.001 -1.264125 -.34908 _cons | 13.12112 4.201755 3.12 0.003 4.685745 21.55649 ------------------------------------------------------------------------------

15

Equivalence of the EC and ADL models

The results imply the equation

Equivalence of the EC and ADL Models

Yt = 13.12 + .58*Yt-1 + 1.32*Xt -.81*Xt-1 + t
The total long term effect/long run multiplier can be calculated as (1.32 - .81)/(.58 - 1) = 1.22 which is equivalent to the ECM estimate. Note, however, that we do not have a standard error for the long run multiplier. Y and X will be in their long term equilibrium state when

Yt = 13.12 + .58*Yt-1 + 1.32*Xt -.81*Xt-1 + t

Our estimate of the contemporaneous effects of X on Y 1.32 units: the same as in the ECM. The long term effect of X on Y at t+1 can be calculated as: 1.32 - .81 = .52 which is equivalent to the .52 estimate in the ECM Deviations from equilibrium are maintained at a rate of 58% per time period, which implies that deviations from equilibrium are corrected at a rate of 42% per time period (.58 - 1).

Y = 30.89 + 1.22X

Error Correction Models

A Flexible Modeling approach
Stationary and Integrated Data Long and Short Term Effects

Engle and Granger two-step ECM versus Single Equation ECM

Importance of Theory Integrated or Stationary Data? Single Equation ECMs avoid this debate. Single equation ECMs dont require cointegration and ease interpretation of causal relationships.

Single equation ECMs and ADL models

Equivalence: ADL models can provide the same information about short and long term effects. Standard error for the long term effects of independent variables is relatively easy to obtain in the single equation ECM

16