Ignoring Y with known X using VECM in R

I am trying to represent time series data (Y) using a different time series (X) as a predictor. X and Y are cointegrated. Y is monthly data from January 2012 to October 2016, and X is from January 2012 to February 2017.

So, I ran VECM as shown in this video: https://www.youtube.com/watch?v=x9DcUA9puY0

Then, to get the predicted values, I converted it to VAR with the command vec2var

following information from this topic: https://stats.stackexchange.com/questions/223888/how-to-forecast-from-vecm-in-r

But I cannot predict Y with known X, as can be done with a function predict

with a linear regression model. Also, I cannot get the simulated Y (Y hat) values.

This is my code:

# Cointegrated_series is a ZOO object, which contains two time series X and Y

library("zoo") 
library("xts") 
library("urca")
library("vars")

# Obtain lag length
Lagl <- VARselect(Cointegrated_series)$selection[[1]]

#Conduct Eigen test
cointest <- ca.jo(Cointegrated_series,K=Lagl,type = "eigen", ecdet = "const", 
                    spec = "transitory")
#Fit VECM
vecm <- cajorls(cointest)

#Transform VECM to VAR
var <- vec2var(cointest)

      

        
        
        
      

    

What I'm trying to use a function predict

in different ways: predict(var)

, predict(var, newdata = 50)

, predict(var, newdata = 1000)

- the result is the same.

Tried using tsDyn

package and newdata

argument in method predict

as he mentioned here: https://stats.stackexchange.com/questions/261849/prediction-from-vecm-in-r-using-external-forecasts-of-regressors?rq=1

Does not work. My newdata is a ZOO object where X-series is from November 2016 to February 2017 and Y-series is NA. Thus, the method returns NA in the forecast:

    # Cointegrated_series is a ZOO object, which contains 
#two time series X and Y from Jan 2012 to Oct 2016. Both X and Y are values.
    # newDat is a ZOO object, which contains two time series 
#X and Y from Nov 2016 to Feb 2017. X are values, Y are NAs.

    library(tsDyn)

    vecm <-VECM(Cointegrated_series, lag=2)

    predict(vecm,newdata = newDat, n.ahead=5)

      

        
        
        
      

    

This is the result:

                                                    Y             X
59                                                  NA            NA
60                                                  NA            NA
61                                                  NA            NA
62                                                  NA            NA
63                                                  NA            NA

      

        
        
        
      

    

For example, this is what I get after calling the predict

whithout argument newdata

:

predict(vecm, n.ahead=5)

                                              Y             X
59                                            65.05233      64.78006
60                                            70.54545      73.87368
61                                            75.65266      72.06513
62                                            74.76065      62.97242
63                                            70.03992      55.81045

      

        
        
        
      

    

So my main questions are:

• How do I now run Y with a known X using the VEC model in R?
• How do I get the simulated Y values ​​(Y hat)?

Also, I couldn't find the answer to these questions either:

3. What to call the Akaike criteria (AIC) for VECM in R?

4. Are there vars and urca packages for F and t statistics for VECM?

UPD 04/10/2017 I slightly edited the question. It is noticed that my problem is related to the problem of "dangling edge", and it is incorrect to call it "forecasting" - it is "forecasting".

UPD 11.04.2017

Thanks for answering!

Here's the complete code:

library("lubridate")
library("zoo")
library("xts") 
library("urca")
library("vars")
library("forecast") 



Dat <- dget(file = "https://getfile.dokpub.com/yandex/get/https://yadi.sk/d/VJpQ75Rz3GsDKN")
NewDat <- dget(file = "https://getfile.dokpub.com/yandex/get/https://yadi.sk/d/T7qxxPUq3GsDLc")



Lagl <- VARselect(Dat)$selection[[1]]

#vars package

cointest_e <- ca.jo(Dat,K=Lagl,type = "eigen", ecdet = "const", 
                    spec = "transitory")


vecm <- cajorls(cointest_e)

var <- vec2var(cointest_e)

Predict1 <- predict(var)

Predict2 <- predict(var, newdata = NewDat)

Predict1$fcst$Y
Predict2$fcst$Y

Predict1$fcst$Y == Predict2$fcst$Y
Predict1$fcst$X == Predict2$fcst$X

#As we see, Predict1 and Predict2 are similar, so the information in NewDat
#didn't came into account.

library("tsDyn")

vecm2 <-VECM(Dat, lag=3)

predict(vecm2)
predict(vecm2, newdata=NewDat)

      

        
        
        
      

    

If it dget

returns an error, please upload my details here:

https://yadi.sk/d/VJpQ75Rz3GsDKN - for Dat

https://yadi.sk/d/T7qxxPUq3GsDLc - for NewDat

About nowcasting

Saying Nowcasting I mean current or previous months projections of unavailable data with currently available data. Here are some links:

Gianonne, Reichlin, Small: Nowcasting: Information Content of Real-Time Macroeconomic Data (2008)

Now-Casting and Real Time Data Stream (2013)

Marcelino, Schumacher: The MIDAS Factor for Nowcasting and Forecasting Using Edged Data: A Comparison of Models for German GDP (2010)