Demand Forecasting : Monte carlo Simulation
Locked
 
|
I am trying to build a demand forecasting model for human resource team. I have thought of using monte carlo simulation method to do it. Is it the right technique for it? Has anyone used it to forecast human capital? What variables should i take into consideration to develop a model? The variables i can think of are team size, number of projects in a month or quarter, growth rate, attrition rate etc. It's a IT company. They are facing high attrition.
Any suggestion is highly appreciated! |
Re: Demand Forecasting : Monte carlo Simulation
|
|
To do forecasting you need to have a regression model of some type. Do you have that?
If you have that, Ryan and Bruce have posted examples of using the Input program command to generate data for a specific model. The key difference between what they did and what you want is that they generated one dataset, you might want to generate 100, 500, or 1000; analyze each; and summarize the results. So that Jon won't have to do this, I'd also like to point out that spss has an add-on forecasting module, which may work depending on your regression model, and I think v23 has an add-on data simulation module. Gene Maguin -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ujjawal Sent: Monday, July 14, 2014 4:01 PM To: [hidden email] Subject: Demand Forecasting : Monte carlo Simulation I am trying to build a demand forecasting model for human resource team. I have thought of using monte carlo simulation method to do it. Is it the right technique for it? Has anyone used it to forecast human capital? What variables should i take into consideration to develop a model? The variables i can think of are team size, number of projects in a month or quarter, growth rate, attrition rate etc. It's a IT company. They are facing high attrition. Any suggestion is highly appreciated! -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Demand-Forecasting-Monte-carlo-Simulation-tp5726700.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 |
|
|
Gene and others: Here's a small simulation example I provided a long time ago that generated 10 datasets of N = 1500: As you can see, generating multiple datasets using the approach in that link simply required an additional LOOP. Ryan
On Wed, Jul 16, 2014 at 9:42 AM, Maguin, Eugene <[hidden email]> wrote: To do forecasting you need to have a regression model of some type. Do you have that? |
|
|
Ryan,
=====================
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 for all this. I have a related question, for which I probably should initiate a new thread, but maybe someone has a solution. I've tried lots of approaches to use Monte Carlo to generate a variable(s) with normal distribution (easy) with the additional
limit to the min and max values. I can't seem to get the syntax which would combine the two. Uniform uses min and max, and normal uses the mean and stddev. Any ideas?? Thanks all. I'll start a new thread if my netiquette is off.
Brian
From: SPSSX(r) Discussion [[hidden email]] on behalf of Ryan Black [[hidden email]]
Sent: Wednesday, July 16, 2014 12:44 PM To: [hidden email] Subject: Re: Demand Forecasting : Monte carlo Simulation Gene and others:
Here's a small simulation example I provided a long time ago that generated 10 datasets of N = 1500:
As you can see, generating multiple datasets using the approach in that link simply required an additional LOOP.
Ryan
On Wed, Jul 16, 2014 at 9:42 AM, Maguin, Eugene
<[hidden email]> wrote:
To do forecasting you need to have a regression model of some type. Do you have that? |
Re: Demand Forecasting : Monte carlo Simulation
|
|
If you are using the simulation procedure
(Analyze > SIMULATION), you can specify all this in the dialog and paste
it.
Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 From: "Dates, Brian" <[hidden email]> To: [hidden email], Date: 07/16/2014 10:55 AM Subject: Re: [SPSSX-L] Demand Forecasting : Monte carlo Simulation Sent by: "SPSSX(r) Discussion" <[hidden email]> Ryan, Thanks for all this. I have a related question, for which I probably should initiate a new thread, but maybe someone has a solution. I've tried lots of approaches to use Monte Carlo to generate a variable(s) with normal distribution (easy) with the additional limit to the min and max values. I can't seem to get the syntax which would combine the two. Uniform uses min and max, and normal uses the mean and stddev. Any ideas?? Thanks all. I'll start a new thread if my netiquette is off. Brian From: SPSSX(r) Discussion [[hidden email]] on behalf of Ryan Black [[hidden email]] Sent: Wednesday, July 16, 2014 12:44 PM To: [hidden email] Subject: Re: Demand Forecasting : Monte carlo Simulation Gene and others: Here's a small simulation example I provided a long time ago that generated 10 datasets of N = 1500: https://listserv.uga.edu/cgi-bin/wa?A2=ind1101&L=SPSSX-L&P=R28980&1=SPSSX-L&9=A&I=-3&J=on&d=No+Match%3BMatch%3BMatches&z=4 As you can see, generating multiple datasets using the approach in that link simply required an additional LOOP. Ryan On Wed, Jul 16, 2014 at 9:42 AM, Maguin, Eugene <emaguin@...> wrote: To do forecasting you need to have a regression model of some type. Do you have that? If you have that, Ryan and Bruce have posted examples of using the Input program command to generate data for a specific model. The key difference between what they did and what you want is that they generated one dataset, you might want to generate 100, 500, or 1000; analyze each; and summarize the results. So that Jon won't have to do this, I'd also like to point out that spss has an add-on forecasting module, which may work depending on your regression model, and I think v23 has an add-on data simulation module. Gene Maguin -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ujjawal Sent: Monday, July 14, 2014 4:01 PM To: [hidden email] Subject: Demand Forecasting : Monte carlo Simulation I am trying to build a demand forecasting model for human resource team. I have thought of using monte carlo simulation method to do it. Is it the right technique for it? Has anyone used it to forecast human capital? What variables should i take into consideration to develop a model? The variables i can think of are team size, number of projects in a month or quarter, growth rate, attrition rate etc. It's a IT company. They are facing high attrition. Any suggestion is highly appreciated! -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Demand-Forecasting-Monte-carlo-Simulation-tp5726700.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== To manage your subscription to SPSSX-L, send a message to LISTSERV@... (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 LISTSERV@... (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 LISTSERV@... (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 LISTSERV@... (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, Jon. That was easy!!!
=====================
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
Brian
From: Jon K Peck [[hidden email]]
Sent: Wednesday, July 16, 2014 12:59 PM To: Dates, Brian Cc: [hidden email] Subject: Re: [SPSSX-L] Demand Forecasting : Monte carlo Simulation If you are using the simulation procedure (Analyze > SIMULATION), you can specify all this in the dialog and paste it.
Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 From: "Dates, Brian" <[hidden email]> To: [hidden email], Date: 07/16/2014 10:55 AM Subject: Re: [SPSSX-L] Demand Forecasting : Monte carlo Simulation Sent by: "SPSSX(r) Discussion" <[hidden email]> Ryan, Thanks for all this. I have a related question, for which I probably should initiate a new thread, but maybe someone has a solution. I've tried lots of approaches to use Monte Carlo to generate a variable(s) with normal distribution (easy) with the additional limit to the min and max values. I can't seem to get the syntax which would combine the two. Uniform uses min and max, and normal uses the mean and stddev. Any ideas?? Thanks all. I'll start a new thread if my netiquette is off. Brian From: SPSSX(r) Discussion [[hidden email]] on behalf of Ryan Black [[hidden email]] Sent: Wednesday, July 16, 2014 12:44 PM To: [hidden email] Subject: Re: Demand Forecasting : Monte carlo Simulation Gene and others: Here's a small simulation example I provided a long time ago that generated 10 datasets of N = 1500: https://listserv.uga.edu/cgi-bin/wa?A2=ind1101&L=SPSSX-L&P=R28980&1=SPSSX-L&9=A&I=-3&J=on&d=No+Match%3BMatch%3BMatches&z=4 As you can see, generating multiple datasets using the approach in that link simply required an additional LOOP. Ryan On Wed, Jul 16, 2014 at 9:42 AM, Maguin, Eugene <[hidden email]> wrote: To do forecasting you need to have a regression model of some type. Do you have that? If you have that, Ryan and Bruce have posted examples of using the Input program command to generate data for a specific model. The key difference between what they did and what you want is that they generated one dataset, you might want to generate 100, 500, or 1000; analyze each; and summarize the results. So that Jon won't have to do this, I'd also like to point out that spss has an add-on forecasting module, which may work depending on your regression model, and I think v23 has an add-on data simulation module. Gene Maguin -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ujjawal Sent: Monday, July 14, 2014 4:01 PM To: [hidden email] Subject: Demand Forecasting : Monte carlo Simulation I am trying to build a demand forecasting model for human resource team. I have thought of using monte carlo simulation method to do it. Is it the right technique for it? Has anyone used it to forecast human capital? What variables should i take into consideration to develop a model? The variables i can think of are team size, number of projects in a month or quarter, growth rate, attrition rate etc. It's a IT company. They are facing high attrition. Any suggestion is highly appreciated! -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Demand-Forecasting-Monte-carlo-Simulation-tp5726700.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 ===================== 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 |