Subscribe to DSC Newsletter

I have to build a prediction model for employee requirement. Please suggest how to go about it. I am thinking of taking 3 quarter's data and by taking the moving average predict it for the 4th quarter. Thsi is so because for every year the resource requirement will be different. Also suggest if some other data mining algorithms can be applied over the emplyee database to achieve something from the data mining view point.

Views: 304

Replies to This Discussion

Hi DataMiner....I'd be happy to help out, as this kind of analysis is high value and high stakes. To be as helpful as possible, it would be helpful for me to know a bit more about the type of business for which you are performing this analysis, the function(s) of the employees in question, and most importantly, the known drivers of employee demand.

As an example, if the business was retail, where demand is driven by retail foot traffic and sales volumes, the staff demand curve will be very seasonal, which might require a different method from the one you described above. On the other hand, if the business is not seasonal, we would want to know what DOES drive the variation in staff demand, and examine the predictive impact of related variables.

Let me know if this is helpful,

Jaime Fitzgerald
President, Fitzgerald Analytics
Architects of Fact-Based Decisions
Thanks Jaime, your thoughts are usefull. Answers to your questions are as follows:- The business is that of a software company. I want to predict the resources required in a quarter for a software company. The function of the employees is typically coding, analysis, design. The driving part depends upon the economical situation of the company , how much budget is allocated for a project. This driving thing results in uncertain demand of resources each quarter.

also let me know which predicting method would be sutiable if it a seasonal type of activity.
And please share your views as to when is the moving average method more applicable in such type of problems.

Should I assume that you are based in U.S and I will have to look out for your replies during my daytime due to the time difference and I being in India.
Hi Jaime,

Please let me know when r u online so that I can try to be online during that period.

Sure! Let me get back to you when I know I will be both online and available to chat. Looking forward!


Jaime Fitzgerald
Hi Jaime,

In the mean time you can go through the answers to the questions in the above communication.

Hi Jaime ,

I am waiting for you.

Sincere apologies Amit -- I have been MUCH slower to reply than I intended, due to being unexpectedly busy helping clients. If the topic is still relevant giving your workflow since we last corresponded, please find me on skype so we can chat and/or talk about it. How about tomorrow morning @ 6:30AM ET?
Jaime i don't have a skype id, so will use google, is it okay with you. 6:30 am at your end means 5:00 pm at my end(IST), so should not be problem. Let me know your google email address.
Great! will send you my gmail address directly and speak with you then.
sent you a "friend request" here, so I can send direct messages. IN the meantime, you can add my gmail account to your contacts, I am at jaime.fitzgerald at gmail

Thanks for accepting my friend that we are connected I was able to direct message you with my gmail account....feel free to chat me when time permits. Looking forward to meeting you and thought-partnering on this or other analytic topics
we can chat here itself


On Data Science Central

© 2021   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service