Stepwise Regression

Recently read this article, courtesy from another fellow analytics enthusiast. This paper discuss about a popular statistical modelling myth, commonly seen in the private sector of Singapore. And that is the stepwise regression.

Based on my observation, most of the modellers in Singapore are actually using Stepwise Regression when modeling linear models. The idea of Stepwise Regression seems to appeal a lot of statisticians in Singapore because it takes variables in and variables out, having a lot of combination of variables in the model and choosing the best combination.

But after reading this paper, it seems like doing stepwise regression, that are certain pitfalls and one of them is the exaggeration of p-value. Now this is very scary, given that it is precisely p-value that we chose a particular variable to be inside the model.

This will be an interesting direction to figure out why that is so given the mechanism/process of selecting variables.

On another note, there is an interesting modelling technique that is inside the paper and that is LARS. What it does is, it chose variables by choosing the next best correlated variables to the error term. It would definitely take up a lot of time to model but am sure it makes a lot of sense to me. But I would feel that a simple correlation table has to be done for all the variables to determine multi-collinearity problem. But my take that such a possibility is very low.

For those that are interested in reading the paper, please refer to the link below.

www.nesug.org/proceedings/nesug07/sa/sa07.pdf

Conjoint Analysis in Marketing

Conjoint analysis has been used in Marketing for the longest while but it seems like it has not taken root here in Singapore. In any case, Conjoint analysis is used in marketing by finding out which features of a product can satisfy the customers more so that companies can improve on these features to gain more customer satisfaction with the least manufacturing costs and thus better profit margins.

It seems like for conjoint analysis, segmentation is necessary to derive more insights. Perhaps a segmentation analysis need to be done first to understand the key customer segments.

http://www.quickmba.com/marketing/research/conjoint/

http://en.wikipedia.org/wiki/Conjoint_analysis_%28marketing%29

http://www.speedback.com/conjoint_analysis.htm

Talent Analytics (HBR)

With reference to HBR Oct 2010, there is an article on talent analytics written by Thomas Davenport, Jeanne Harris and Jeremy Shapiro. In this article they discussed about how companies in the US are doing Talent Analytics and the six ways in which analytics are being used in Human resource management.

1) Human capital facts - a single version of truth regarding employee's performance and enterprise level data such as head count, turnover and recruiting. (Net Promoter Score)
2) Analytical HR - segment HR data to identify turnover intervention, matching performance with dept objectives to identify the need for intervention
3) Human capital investment analysis - helps an organization to understand which actions have the greatest impact on business performance. Understand which metrics for instance quality of life, supervisor effectiveness, have an impact on its performance.
4) Workforce forecasts - Staffing up key growth areas and identify knowledge management risks for retiring employees.
5) Talent Value Model - To calculate what employees value most and create a model to boost retention rate. The model can be used for designing personalized incentives, assess whether to match competitor's recruitment offer or decide to promote someone. Also to determine is there is a need to change the scope of work.
6) Talent Supply Chain - optimizing work schedule, forecasting volume through call-centre. Strong need for high quality data.