AR(1) model in SPSS
Locked
 
|
Can someone plz help me with this:
I am trying to calculate mean reversion for daily interest rates/volatilities: I have my model in the form: R(T) = a + b * R(T-1) + E(T) where R(T) is the daily rate at time T. now I want to implement an AR(1) model here to find out my estimates of a and b. Can someone plz tell me how to do this in SPSS? I do have the time series add in, but I can't seem to figure out where AR goes. I have tried searching the help on SPSS too. It just talks about AR(1) but not how to actually do it in SPSS. I trhink I have to go to Analyze -> Time Series ->Create models but I only see ARIMA not AR. Are they related? Any ideas would be great. THanks. |
|
|
AR(1) is the same thing as ARIMA(1,0,0).
In the Time Series Modeler, select ARIMA as the method. Click Criteria. Enter 1 in the AR field (nonseasonal) Leave the others at zero. You may want to check Parameter Estimates on the Statistics tab. HTH, Jon Peck -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of jimjohn Sent: Thursday, November 27, 2008 2:09 PM To: [hidden email] Subject: [SPSSX-L] AR(1) model in SPSS Can someone plz help me with this: I am trying to calculate mean reversion for daily interest rates/volatilities: I have my model in the form: R(T) = a + b * R(T-1) + E(T) where R(T) is the daily rate at time T. now I want to implement an AR(1) model here to find out my estimates of a and b. Can someone plz tell me how to do this in SPSS? I do have the time series add in, but I can't seem to figure out where AR goes. I have tried searching the help on SPSS too. It just talks about AR(1) but not how to actually do it in SPSS. I trhink I have to go to Analyze -> Time Series ->Create models but I only see ARIMA not AR. Are they related? Any ideas would be great. THanks. -- View this message in context: http://www.nabble.com/AR%281%29-model-in-SPSS-tp20725617p20725617.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
|
|
Thanks very much Jon! I just have a follow up question:
I tried running an AR(1) in SPSS. My variable is CADSWAP and I want to find my AR(1) model parameters for this data. So I just entered CADSWAP into my dependent variable box and ran the AR model. The output I get from this for my parameters is pasted below, so I assumed my equation would be: Y(T+1) = .029 + .999 * Y(T) + E(T) But, I saved the predicted values, and I compare this equation with the actual estimates of the model, and it doesn't work out. Just wondering what am I doing wrong, am I not interpreting the SPSS output correctly? Or, do I need to enter in Y(T+1) as the dependent variable and Y(T) as the independent variable? (because in that case, I keep getting an error saying there are missing values). Thanks a lot for your help! ARIMA Model Parameters Estimate SE t Sig. CADSWAP1-Model_1 CADSWAP1 No Transformation 0 Constant .029 .006 4.729 .000 AR Lag 1 .999 .001 1566.191 .000
|
|
|
are you using spss trends ? try exporting the model into xml and see.
ajay Bob Dole - "The internet is a great way to get on the net." On Tue, Dec 2, 2008 at 10:03 PM, jimjohn <[hidden email]> wrote: > Thanks very much Jon! I just have a follow up question: > I tried running an AR(1) in SPSS. My variable is CADSWAP and I want to find > my AR(1) model parameters for this data. So I just entered CADSWAP into my > dependent variable box and ran the AR model. The output I get from this for > my parameters is pasted below, so I assumed my equation would be: > Y(T+1) = .029 + .999 * Y(T) + E(T) > > But, I saved the predicted values, and I compare this equation with the > actual estimates of the model, and it doesn't work out. Just wondering what > am I doing wrong, am I not interpreting the SPSS output correctly? Or, do I > need to enter in Y(T+1) as the dependent variable and Y(T) as the > independent variable? (because in that case, I keep getting an error saying > there are missing values). > > Thanks a lot for your help! > > > > > ARIMA Model Parameters > Estimate > SE t Sig. > CADSWAP1-Model_1 CADSWAP1 No Transformation 0 Constant .029 .006 > 4.729 .000 > AR Lag 1 .999 .001 > 1566.191 .000 > > > > > > > > > Peck, Jon wrote: > > > > AR(1) is the same thing as ARIMA(1,0,0). > > In the Time Series Modeler, select ARIMA as the method. > > Click Criteria. > > Enter 1 in the AR field (nonseasonal) > > Leave the others at zero. > > > > You may want to check Parameter Estimates on the Statistics tab. > > > > HTH, > > Jon Peck > > > > -----Original Message----- > > From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of > > jimjohn > > Sent: Thursday, November 27, 2008 2:09 PM > > To: [hidden email] > > Subject: [SPSSX-L] AR(1) model in SPSS > > > > Can someone plz help me with this: > > > > I am trying to calculate mean reversion for daily interest > > rates/volatilities: > > > > I have my model in the form: > > R(T) = a + b * R(T-1) + E(T) > > > > where R(T) is the daily rate at time T. > > > > now I want to implement an AR(1) model here to find out my estimates of a > > and b. Can someone plz tell me how to do this in SPSS? I do have the time > > series add in, but I can't seem to figure out where AR goes. I have tried > > searching the help on SPSS too. It just talks about AR(1) but not how to > > actually do it in SPSS. I trhink I have to go to Analyze -> Time Series > > ->Create models but I only see ARIMA not AR. Are they related? Any ideas > > would be great. THanks. > > -- > > View this message in context: > > http://www.nabble.com/AR%281%29-model-in-SPSS-tp20725617p20725617.html > > Sent from the SPSSX Discussion mailing list archive at Nabble.com. > > > > ===================== > > To manage your subscription to SPSSX-L, send a message to > > [hidden email] (not to SPSSX-L), with no body text except the > > command. To leave the list, send the command > > SIGNOFF SPSSX-L > > For a list of commands to manage subscriptions, send the command > > INFO REFCARD > > > > ===================== > > To manage your subscription to SPSSX-L, send a message to > > [hidden email] (not to SPSSX-L), with no body text except the > > command. To leave the list, send the command > > SIGNOFF SPSSX-L > > For a list of commands to manage subscriptions, send the command > > INFO REFCARD > > > > > > -- > View this message in context: > http://www.nabble.com/AR%281%29-model-in-SPSS-tp20725617p20794415.html > Sent from the SPSSX Discussion mailing list archive at Nabble.com. > > ===================== > To manage your subscription to SPSSX-L, send a message to > [hidden email] (not to SPSSX-L), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSX-L > For a list of commands to manage subscriptions, send the command > INFO REFCARD > ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |