A unit root is a property of a non-stationary time series which can lead to spurious regressions and wrong inference.

learn more… | top users | synonyms

0
votes
0answers
12 views

Cointegration test of stationary variables with structural breaks

I am trying to implement test of cointegration (Johansen trace and eigenvalue) between 2 variables that are stationary at first difference but after Zivot-Andrews test of unit root I found they have ...
2
votes
0answers
30 views

R: Stationarity/non-Stationarity: implementing a solution

I have an Augmented Dickey Fuller Test in R on leg_totalbills that shows I can not reject the null hypothesis: Unit Root. The ...
1
vote
1answer
24 views

How does trend stationary recovers from shocks in long run?

I was trying to understand difference between drift and trend wherein I came across concepts of unit roots and trend stationary. (I haven't read any books on time series, just going through web). ...
0
votes
0answers
29 views

Optimal lag selection for panel unit root test PESCADF and CIPS

I am running second generation panel unit root test in stata, CIPS with "multipurt" and PESCADF with "pescadf", but I would like to know if there is a way to know the optimal lag selection, and if I ...
0
votes
0answers
7 views

xtunitroot rho estimation [closed]

I am performing the following unit root tests on panel data: Levin-Lin-Chu xtunitroot llc y, lags(1) trend demean Augmented Dickey-Fuller xtunitroot fisher y, dfuller lags(1) trend demean Based on ...
1
vote
0answers
4 views

Is it ok to use Panel Unit Root Tests for all currency pairs in a sample? (Not one common base currency)

I have a dataset containing the real exchange rates for all country pairs within the OECD. Now I would like to test stationarity using one of the many panel unit root tests. However, reading Pesaran, ...
3
votes
1answer
39 views

How is the augmented Dickey–Fuller test (ADF) table of critical values calculated?

Could you please explain in simple terms how the table of critical values for the augmented Dickey–Fuller (ADF) test is created?
0
votes
1answer
35 views

logarithm and absolute value in returns of stocks

I'm interested in model a GARCH for a serie. The original serie is $y_t$ (price index of a Stock Market), which has a unit root. So I created the returns: $x_t = \ln(y_t) - \ln(y_{t-1})$. Now, I'm ...
1
vote
0answers
27 views

Comparing unit root tests

When comparing the results of different unit root tests (in this case DF-GLS test with and ADF test), should I keep the lag length fixed for both tests, or am-I allowed to use predefined/suggested ...
3
votes
1answer
33 views

Autoregressive model for time series with structural breaks

I'm using a structural break model (threshold model or regime switching model) to examine the dynamics of a time series. The ADF test shows that the series has a unit root. Right now I'm regressing ...
1
vote
1answer
32 views

Phillips-Perron Unit Root Test with exogenous breaks

I would need to do a Philips-Perron test in R, while controlling for an exogenous break. I found the Philips-Perron test in the package tseries (PP.test), but can't find (contrary to the CADF test) ...
0
votes
1answer
53 views

Choosing the maximum lag length in the augmented Dickey-Fuller test

I have a question regarding how to choose the maximum lag length in the augmented Dickey-Fuller test using the "urca" package in R. I want to perform the ADF test on the daily price of a stock index ...
0
votes
0answers
24 views

What do you think of this correlogram?

do you think this weekly data is stationary? Unit-root test indicates rejects null of non-stationary (rejects null of unit root). Thanks for your input.
0
votes
0answers
16 views

Dickey fuller/ unit-root test controversy

The regression equation of dickey fuller test is defined as: $$\Delta y_t = py_{t-1}+e_t$$ Trends and constants may also be added.The test statistic can be derived in the following way: $$y_t=\beta ...
0
votes
0answers
13 views

Granger causality test for integrated variables: to difference or not to difference? [duplicate]

Given a pair of I(1) variables, should I difference them before conducting the Granger causality test? Additionally, which lag would I use to check the causality?
1
vote
1answer
45 views

Is a trend-stationary variable I(1) or I(0)?

I am trying to do cointegration analysis between two variables. I first used the standard Dickey-Fuller and Phillips-Perron tests; they concluded my variables were I(1). I then did cointegration and ...
2
votes
1answer
54 views

Does using difference transformation lead to bias? (Levels vs differences regression)

Consider the model estimated in levels (also assume this is the true population model): $$y_t = x_t\beta + e_t$$ As usual we have the dependent variable $y$, independent $x$, the error term $e$, and ...
1
vote
0answers
23 views

Critical values for Levin & Lin (1992) unit root test with lags

I am testing for the presence of unit root in a panel data with $T=12$ and $N=60$. I am using the Levin & Lin "Unit root tests in Panel data: asymptotic and finite-sample properties" (1992) ...
0
votes
0answers
41 views

ADF Test Variation between tseries::adf.test in R and Matlab adftest

I'm having an issue getting the p-values out of the ADF Test to match between the tseries package in R and the output in Matlab. Using the example from the Matlab help located in the help files ...
0
votes
0answers
23 views

How to code dummy variables for structural breaks in VAR

This question is really 2-in-1: 1) How do I code dummy variables for the following series that has 2 structural breaks in trend; an initial upward trend, then a much flatter upward trend, then ...
0
votes
1answer
27 views

Unit root tests ambigous - is time series stationary?

I am testing a time series (quarterly) for stationarity. However, using the KPSS test, the ADF test and PP test, I get different results (ADF and PP reject non-stationarity, KPSS rejects stationarity, ...
0
votes
0answers
19 views

More than one unit root

I'm wondering how to detect the existence of more than one unit root. I've tested the non-stationnarity of my serie using the strategy of test of Dickey-Fuller (joint hypothesis: Fisher), and I detect ...
0
votes
1answer
22 views

Are we using transformed or original series to fit a time series model?

I would like to know whether one can use the transformed series to fit a model or not. For example, fitsereies=auto.arima(differenced_series) or ...
1
vote
0answers
20 views

What would be an apporiate model for regressing multiple non-stationary time series data?

I have non-stationary time-series data (stationarity tested using ADF Test) for variables such as stock market returns, money supply, interest rates, exchange rate, inflation,etc. and I want to study ...
2
votes
1answer
57 views

Criticism of augmented Dickey-Fuller test

I am looking for literature that suggests the augmented Dickey-Fuller (ADF) test is not completely accurate, or in general criticizing the test.
1
vote
0answers
29 views

Testing for Unit Roots with Quadratic Trends in Stata

I understand that Dickey-Fuller test could test for a unit root with drift and deterministic time trend. $$ \nabla y_t = a_0+a_1t+\delta y_{t-1}+u_t \ $$ What are the tests for unit root with ...
0
votes
0answers
34 views

Choosing specification of the augmented Dickey-Fuller test

I am testing whether my time series data is stationary or not. But I do not know what to include in the test equation of the augmented Dickey-Fuller (ADF) test (Trend, Trend and Intercept, or None). ...
0
votes
0answers
17 views

Relationship between T, the Adjustment Rate, and Power for an Autoregressive Unit Root

So I have a question related to autoregressive unit roots. This picture shows a graph for c and n with power = .9 in blue and power = .8 in red. Any ideas what could be used to fit this graph or ...
4
votes
3answers
93 views

Difference between $y_t = \alpha + \beta t$ and $y_t = y_{t-1} + \beta$

Would someone mind walking me through the differences between: \begin{align} y_t &= \alpha + \beta t \\ &\& \\ y_t &= y_{t-1} + \beta \end{align} as well as between \begin{align} ...
0
votes
0answers
17 views

Interpret the result of augmented Dickey-Fuller test

As per the title, can anyone help me with this result? I have three more to analyze so hope that can learn it.
1
vote
0answers
15 views

Unit root: differencing vs. using % change vs. taking ln

Let's say your trying to forecast a country's GDP with an ARMA model. Obviously the original series will contain a unit root (at least most of the time), so which series would you use the model on? Is ...
3
votes
0answers
34 views

Why is cointegration found without unit root?

I'm working on the multiple price series data to look for the long run relationships. DF-GLS tests for unit root are rejected although the series show some trends and seem stationary after ...
1
vote
2answers
72 views

Johansen cointegration test: interpret the result

I am trying to develop a stationary time series with 6 variables in Matlab. Can anyone tell me whether the 6 variables are cointegrated? The results I got are:
0
votes
1answer
34 views

Why does this condition on AR coefficients imply a unit root?

Consider an AR($q$) process: $$ X_t-\theta_1 X_{t-1}-...-\theta_q X_{t-q}=Z_t $$ where $Z_t$~$WN(0,1)$. Why does the following condition on the process's coefficients imply a unit root: $$ ...
0
votes
1answer
43 views

Unit roots Dickey Fuller test question

I've been searching in bibliography about this test applied to an ARMA(p,q) model, and find out that every single book states the null hypothesis as "1 is a root of the operator". I was wondering if ...
0
votes
1answer
37 views

Augmented Dickey-Fuller test and Differencing (R)

I'm using the ADF test to check for stationarity of two variables, using the ndiffs function in R. ...
0
votes
1answer
42 views

Is there an unambiguous stationarity test for time series?

It seems to me that the time series plot, the correlograms (ACS, PACS) and the "autocorrelation check for residuals test" can all be subject to interpretation. (I am using SAS 9.4) Is there an ...
2
votes
1answer
68 views

Testing for drift in random walk

How would you test whether a unit root I(1) process has a statistically significant constant? To my understanding, the AR(1) regression in the levels is statistically spurious, and hence we cannot ...
0
votes
0answers
26 views

Some, but not all, panels seem to cointegrate

I am trying to estimate the long-run relationship between equity prices and bond default risk measures using a 250*250 panel of firms. There is strong theoretical evidence that there is a relationship ...
0
votes
0answers
19 views

Second Differences and Quadratic Trends

I know that any second differenced model has a quadratic trend, but how do you know if its best modelled with stationary or non stationary deviations around this?
0
votes
0answers
144 views

Levin Lin Chu Test for unit root testing in panel data

I want to perform the Levin-Lin-Chu test for unit roots on my pension fund dataset. I have observations: T=10 N=15. The y variable is the portfolio risk of a pension fund in t It is tested if the ...
1
vote
1answer
45 views
0
votes
0answers
14 views

Can i check unit root of series having missing values with eviews or any other software?

I am trying to find the impact of selected expenditure on selected economic growth determinants for 6 developing countries from 1990 to 2013 . Some of the data in the series is missing. So, i want to ...
0
votes
0answers
37 views

Is my Levin Li Chu t test statistic for unit root test in panel data too high?

Method Levin, Lin & Chu t* Statistic -308.914 Prob.** 0.0000 Sample details: Null Hypothesis: Unit root (common unit root process) Series: LMCAP Date: 11/18/15 Time: 12:36 Sample: 2009 ...
1
vote
1answer
57 views

R: Access/store optimal number of lags from unit root test [closed]

I am testing several variables for unit roots via the ur.df (from urca package) and CADFtest ...
0
votes
1answer
100 views

Conflicting results from $p$-value and $t$-value: Should I ignore the $p$-value in the ADF test?

I'm pretty new to the concepts of stationarity/cointegration. I am using the "urca" package in "Rstudio" to run my tests. I have been trying to run cointegration tests, but the frustrating thing is ...
10
votes
3answers
418 views

Nice example where a series without a unit root is non stationary?

I've seen several times people reject the null in an augmented Dickey-Fuller test, and then claim that it shows their series is stationary (unfortunately, I cannot show the sources of these claims, ...
2
votes
0answers
53 views

Implementation of Dickey-Fuller test in R [closed]

I'm working with Dickey-Fuller test. The steps are as follows: Step 1: Starting with model 3, test the significance of time trend. If it isn't significant, then go to step 2. If it is, then test the ...
2
votes
1answer
139 views

Signs on independent variables change in VECM results

I am estimating a VECM model and it was determined by the trace tests that there are two cointegration equations. From this, I proceeded to estimate the β under this restriction that provides the ...
1
vote
1answer
113 views

ADF test for Unit Root, should you select a trend if the trend changes once?

Let's say you have a variable with a time series going back ten years. In the first 5 years, it clearly trends from 5 to 10. And, in the next 5 years, it trends downward from 10 back down to 5. ...