(EViews10): ARDL and 3-Ways Causality Checks (2) #ardl #causality #granger #wald #boundstest

plus we have here around in my second series on air LDL and threeways Casali two checks please before you watch this one make sure you watch the first video where I explain what causality is all about the different definitions and how you come frame is within an LDL module in this video I will only be discussing the first two checks out of the three causality checks this is because I don't want to run all the three checks in one video I really want you to get it you understanding so that you can give me your feedback after you have practiced on your own the tour check will be covered in my third video where I also talk about Diagnostics checks so here in a views I have the group data domestic credit really interest rate and the log of investments I have done stationarity tests and I've also determined the optimal logs if you don't know how to assess for stationary C or determine optimal logs please click on the eave use playlist in the video link below and familiarize yourself with the steps my videos are very simple to follow just load in your data and do exactly what I've done I've also given simple interpretations to all my video tutorials so having said that let us go too quick and I'm going to take domestic queries as a first dependent variable after which I do the same for real interest rates and the log of investments so I go so quick estimates equation and I list the variables I'm changing metal from least squares to LDL and in the a IDL interface I change from cage to took history which is on restricted constants on nutrient the maximum lacks what dependent variable is one what those for the regressors I'm going to use two because the one for real interest rate is locked to everything looks okay then I click OK so on the screen now is a result for domestic credit growth you can see here and you can see the method is a our deal from the output the first check is that you can use the statistical significance of the regressors to also infer causality let's look at the real interest rate this is the coefficient and we consider the p-value is above 0.05 is 18.9 for percents so clearly this is equal to zero it is not significant for the log of investment this is the coefficient point three three and we can see the probe value is four point five percent which is below five percent so this is statistically significant so from these two regressors we can infer that the log of investment has a short-run cosine be face on domestic credit growth the second way to check is by using the world test so we go to view we maneuver to question Diagnostics and we go to wall test you click that now before you perform the wall test you should know that if use our science non bring to the coefficients so the lag or DC growth takes coefficient one really impressed your coefficients test coefficient through this test question tree and the constant is question four so we begin by testing the coefficient of the real interest rates what I chose any causal influence even though we have seen from here there is no significant polettis test anyway so we have C 2 equals zero now we can see the results is not surprising this is the T statistics that we go from the regression and this is the F statistic city chi-square so from here we can see there is no short run causality from real interest rates to domestic credit go let's test for investment using the word test we go to view coefficient Diagnostics manova to wall test so here we type in C 3 because that is a number for the coefficients of logo of investments so C 3 equals 0 so here we can see the F statistics is an agreement with its history sticks at 5% level we can also see the chi-square is also significant at a 5% level so we can also conclude here that the log of investment has a short-run coastal effects on domestic credit same thing for real interest rate as a dependent variable we'll check my tour here to a RDL we trade from case to to case 3 the maximum luck for the real interest rate is - or that regress was one ever that thing looks ok ok this is the outcome for the real interest rates as a dependent variable you can see here and you can see a Rd allows the modal from the regressors coefficients we can see that the first leg of domestic credit is significant at a 1% level and the first leg of investment is also significant at the 5% level so we can say that I'll shorten causality from the first leg of the massive credit and the first leg of investment to really interest rate that is one way to check for short-run cost ality now the second way is by subjecting knees to the world tests like I showed you before let's do the same for the real interest rate we go to view and over to coefficient Diagnostics and we click on wall test so let us give it the appropriate numbering this is number 1 2 3 4 5 6 so we are testing together coefficient 3 and 4 because those are the coefficients attributable to domestic credit so we are testing C 3 C 4 equal 0 remember coefficient C 3 and C 4 are the coefficients of domestic credit so make sure you have put in the appropriate coefficients click ok so this is the outcome of the war chest so the F statistics on ich heisse quay shows that the statistical significance from domestic credit to the real interest rates so we can say here that domestic credit causes the real interest rates in the short run he'll finally let's do the same for the love of investments asked the dependent variable so we go so quick maneuver to estimate equation and type in the variables remember to change your method from Li square so AR DL and again I'm changing to history the dependent variable in this case takes Lag the regressors - laughs because of the real interest rate everything looks okay I click okay so yeah again is the outcome for the logo of investment you consider where as a dependent variable and the method is a RDL so let's take a look at the regressors coefficients I can see here that domestic credit is significant at the 5% level and the second leg of the real interest rate is significant at the 5% level so that is one way to check for short-run causality in the logo of investment equation so now let's go to what test to see whether there's going to be a confirmation or a digression so we go to view coefficient Diagnostics the world test always remember that eviews are science known brings to all the coefficients so that of domestic ready GRU takes coefficient true so c2 equals zero so here we can see that the basic credit has a short-run co-sign influence on the log of investments the p-values are significant at the 5% level so let's do the same thing for the real interest rates we are going to test coefficient three four five because those are the coefficients that are related to the real interest rates we go to view question Diagnostics war test so c3 c4 c5 concededly spelt out C 3 cos C 4 equals C 5 equals zero okay so here we can see that it really interest rates does not have any causal effects in the short run on the log of investments neatly put on the table is the outcome of the two tests that we just did in the domestic credit growth our equation investment is significant and using the word test it's a confirmation same thing when we did the real interest rate regression the first leg of the messy credits and the first leg of investment was significant and what we subjected our to test the outcome is that domestic credits and investments causes real interest rate in the log of investments equation domestic credits is significant and the waters also showed a domestic credit Kosice investment so you can see a similarity between the waters and the T statistics is just a confirmation of the order so we can always come - here - there's a unidirectional causality from the messy credits - really interest rate how do we come to that conclusion if you look at the domestic credits equation there is no influence coming from really interest rates but when you look at your real interest rate equation you can see a causal relationship from domestic credit to really interest rate so that's how we're able to infer a unidirectional causality same thing for investment - real interest rates there is a unidirectional causality from the log of investments to the real interest rates but no causality from real interest rates to investment so is a unidirectional relationship however investments and domestic credit growth exhibits bi-directional causality in the domestic credit growth equation investments exhibits a causal influence or needs likewise in investments regression domestic credit exhibits a causal inference all of these in the short run so like I said using causality checks in your paper or in your manuscript on your dissertation and reaches your walk it brings out a lot of information to your audience and to your readers in conclusion just like I showed you earlier on the T statistics and the F statistics we always indicate short-run causal effects between or among the variables and each of these tests can always serve our robustness for the order you can always use one to validate the other one again if you need more references on how to run causality checks read up this following test books and also sauce order John house howdy it's good to have you please don't go away stay with me in my next video where I'll show you how to run the tour check and also perform some Diagnostics if you have not subscribed to my channel please do so and let me have your feedback thank you

Channel: CrunchEconometrix

Views: 8,577

Rating: 4.9607844 out of 5

Keywords: ardl causality test eviews, granger causality eviews, granger causality test, granger causality in eviews, how to estimate granger causality, how to estimate granger causality eviews, ardl granger evi

Id: zs1mL9pHPRg

Length: 10min 55sec (655 seconds)

Published: Wed May 16 2018