Questions in topic: "prediction"
https://developer.ibm.com/answers/questions/topics/single/192488.html
The latest questions for the topic "prediction"Using a partition in a CHAID model
https://developer.ibm.com/answers/questions/512288/using-a-partition-in-a-chaid-model.html
We are new users of SPSS Modeler and are looking at 3 years' worth of survey data in a CHAID model to predict withdrawal from college. Our dataset is 6300 - is this too small to run a partition? Our accuracy increases by 1% with a partition but the predictive factors change depending on whether or not we use a partition.modelerpredictionchaidFri, 19 Jul 2019 11:38:49 GMTaslynicholsHow can I use a Chaid model for prediction (broadcasting)
https://developer.ibm.com/answers/questions/510713/how-can-i-use-a-chaid-model-for-prediction-broadca.html
Hello,
I am able to successfully build Chaid Model using SPSS 18.1.1; purpose of such model is to predict the flag variable Renew/Not Renew that tells if customer renews or not renew a product licence. Now I would like to use such model to predict the target values for the next months data .
I have tried to do the Generate Modeling Node and that indeed created a new node that should be my model ; than I expect I need to connect the new input data (that has the new data and no the target as I need to predict it) and all the nodes that I use to transform the data (i.e Reclassify ,Deriver or Filter Node) but any time I try to run this new flow it complains that he cannot that a target field is required.....
How can I just re-use the model I have just created to do a prediction of the target with some different data than the one I used to train and so create the model? ThanksmodelerpredictionchaidTue, 02 Jul 2019 14:16:09 GMTMarinucciSpatial-temporal prediction
https://developer.ibm.com/answers/questions/500737/spatial-temporal-prediction.html
I'm trying to use spatial-temporal prediction (geospatial modelling wizard) in order to create a time series model with geospatial data: grocery shops and population distribution according to different censuses.
- Adding maps to the model - done
- Data Sources: done and validated (Context Data: Population per parish between 1987 and 2018 and Data fromPoint Density - grocery shops between 1987 and 2018 - with latitude and longitude).
- Assigning fields: Time Fields (Date format; the same for Context Data and Point Density); Predictors - Population.![alt text][1]
- Settings for building the model and the output - by default. nothing has been changed. Time interval "Periods" increment by "1".
- Run Model - FINISH.
**However, it fails to execute the model with the following warning: "Subcommand DATASET 1: the type for TIMEINDEX variable Data should be interger"**
I kindly ask for support.
![alt text][2]
[1]: /answers/storage/temp/27630-1.png
[2]: /answers/storage/temp/27631-2.pngspssstatisticspredictionWed, 10 Apr 2019 11:54:14 GMTdiogogsilvadiscriminant analysis - predicted correct classification by chance
https://developer.ibm.com/answers/questions/478632/discriminant-analysis-predicted-correct-classifica.html
Hello.
I am running a Discriminant Analysis to see if the age/sex class of a vocalizing animal can be predicted based on the acoustic properties of it's vocalizations. I have seen in a number of publications that use a DFA for this purpose. In their results, they report the percentage of cases correctly classified by the DFA, as well as whether or not this percent is great than what would be expected by chance (with an associated p value). In the Discriminant analysis output, were can I find the percentage of cases that I would expect to be correctly classified by chance alone? Is it included in the output? Where can I find the test comparing the percent correctly classified to the percent expected by chance?
Thank you in advanceclassificationpredictivecontent+classificationdiscriminantpredictionSat, 03 Nov 2018 01:23:09 GMTcjdubreuPrediction nugget export to R or SQL
https://developer.ibm.com/answers/questions/416794/prediction-nugget-export-to-r-or-sql.html
I have a prediction nugget and would like to export its logic to R language. Then in R I would like to pass it a dataset and calculate the outcome. How can I do this? I'm asking this because I want to prove people that running the code in R gives you a similar outcome probability.exportrpredictionTue, 05 Dec 2017 06:06:14 GMTJorgeLopez13Spss Modeler 18: formula for prediction model
https://developer.ibm.com/answers/questions/416526/spss-modeler-18-formula-for-prediction-model.html
Spss Modeler 18: I have a prediction golden nugget from c5 model, and I know that I can predict new datasets by linking it to the golden nugget, but is there (somewhere) the formula the model used to calculate its probability outcomes? Xfc column?modeler18formulapredictionMon, 04 Dec 2017 06:48:28 GMTJorgeLopez13How to do predictions using SPSS
https://developer.ibm.com/answers/questions/416478/how-to-do-predictions-using-spss.html
I currently have a file with the following variables
Time
x1, x2, x3, x4, x5 (x1 - x5 are correlated)
How do I use spss to predict the possible future value of x1?spsshow-tohowpredictivepredictive-analyticspredictionSun, 03 Dec 2017 04:17:00 GMTMangosSPSS Multiple Variable Regression and Prediction
https://developer.ibm.com/answers/questions/292568/spss-multiple-variable-regression-and-prediction.html
Hi, I'm desperately trying to understand how to do a multiple linear regression and predictions. I have lots of data, potentially 7 parts to an independent variable (which is basically one question with 7 parts to the question) then another independent variable that again has about 5 parts and then the dependent variable also has about 5-6 parts to it.
I don't know how to show relationships between the independent and dependent variables in this situation. I had a look online for a one-day (or half-day) training course but couldn't find one. I've got 4 weeks left before my MBA dissertation needs to be handed in, so it is quite urgent if anyone can possibly help me, please? Up until this point, I haven't done any statistical work. I have built SPSS with all of the variables, am just stuck at this analysis stage.
Many thanks
Amandamultiple dataspredictionregressionvariablesTue, 02 Aug 2016 19:47:21 GMTAmandaJGHow do avoid "ArrayIndexOutOfBoundException" error while using Predictive Analytics API in Bluemix
https://developer.ibm.com/answers/questions/268411/how-do-avoid-arrayindexoutofboundexception-error-w.html
I am trying to use Predictive Analytics API in Bluemix with the model given by the Data Analyst, but i always get an error as shown below.
{
"flag": false, "message": "Failed to score risk_calc, msg=null, details:java.lang.ArrayIndexOutOfBoundsException\n\tat com.ibm.spss.internal.PredictiveScorerCSPImpl$OutputStore.getString(PredictiveScorerCSPImpl.java:198)\n\tat com.ibm.spss.blackbox.score.resource.ScoreUtil.score(ScoreUtil.java:190)\n\tat com.ibm.spss.blackbox.score.resource.ScoreResource.scoreWithInputdata(ScoreResource.java:69)\n\tat sun.reflect.GeneratedMethodAccessor102.invoke(Unknown Source)\n\tat ..........
}
But this model runs on the IBM SPSS modeler and gives required output. Also the scoring branch used is based on Text Mining Model. So I think we are missing a link between the IBM SPSS modeler and using it with Bluemix API. Any suggestions on how to resolve this error? Thanks!spssanalyticsibmcloudpredictive-modelingspssscoringpredictionWed, 04 May 2016 06:42:50 GMTGirishSortur