Could anyone help? regards Sanil
>>> Subject: Re: Query on multiple regression Dear Mr. Sanil: You may post your query to the new group [EMAIL PROTECTED] There maybe some one who can answer this question without doing much homework. All the best Krishnamoorthy ----- Original Message ----- From: GMIPDC <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Thursday, March 14, 2002 7:02 AM Subject: Query on multiple regression > Dear Sir > Greetings from India! > > I am working at the production plant of General Motors India in Gujarat. > > Recently i had the opportunity to visit your homepage on the internet. > The reason was that we are facing a theoretical problem in our operations > here. which might need a treatment based on multivariate regression > > was wondering if you could suggest any Algorithm that could be applied > here, > or if any of your students is interested in collaborating with us. > we surely cannot match your competency levels in Statistics > > my work is concerned with managing the SPARE PART INVENTORY. > > we have around 7000 products that are stocked.. and we distribute it > throughout the indian aftersales market. > > GM(Opel) has the SAP Enterprise resources planning system installed over > here.. and we unfortunately dont > utilise its full analytical capabilites > > some of the parts out of the 7000 have erratic demand pattern... some have > trend.. some are cyclic.. and so on > > since the data is little voluminous, i developed a simple package in VB > for > -forecasting demand > -replenishment policy > > forecasting is done for each product,through the best fit of around 14 > different strategies.. from average to wtd average.. exponential > smoothing.. > seasonality.. etc etc > (ill send u a documentation of the software if you feel interested) > > what i was wondering was.. > how to regress the demand pattern data of individual spare parts.. > to data like say > - car population > - no. of service visits to the workshop > - warranty period > and so on.. the variables that might affect are upto us to discover.. > > i dont have much theoretical knowledge about multivariate regression.. > > was wondering if anyone over there has developed an Algorithm which can be > used or modified for such analyses > > warm regards > P.P.Sanil > Spare Parts Distribution Centre > General Motors India (p) Ltd > Chandrapura Industrial Estate > Halol, Gujarat, India > ph: +91 2676 21000 etn 622 > > >>>> > > My research interests include multivariate analysis, statistical inference, > inferences with missing data, statistical calibration, > meta analysis, and designing software using Visual C++. I worked for my > Ph.D. in the area of estimating mutlivariate normal > parameters in a decision theoretic setup. I published several articles on > estimation of normal covariance matrix, precision > matrix, mean vector and eigenvalues. I also contributed in developing > inferential procedures about normal mean vector, > covariance matrix, and generalized variance with incomplete data. > > More recently, my research has focused towards the following areas of > statistics: (i) multivariate calibration, > (ii) statistical methods relevant to occupational exposure and pollution > data, (iii) neurometrics, and > (iv) small sample inference for discrete distributions. > > I am currently collaborating with Thomas Mathew, UMBC, in developing small > sample statistical methods appropriate > for analyzing workplace exposure data, and assessing workers exposure to > workplace contaminants. This is a joint project > funded by the National Institute of Occupational Safety and Health (NIOSH). > This research involves work on lognormal > distribution, tolerance factors for one-way ANOVA random effects model, and > calibration methods. > > I have collaborated with medical scientists on the design of sampling, and > analyzing brain wave data, and worked > with engineers and technicians on the analysis of drilling data and > failure data. > > > K. Krishnamoorthy > Professor > (Ph.D. 1985, IIT-Kanpur, India) > Department of Mathematics > University of Louisiana at Lafayette > Lafayette, LA 70504-1010 > Phone: 337-482-5283; Fax: 337-482-5346 > E-mail: [EMAIL PROTECTED] > > > > > ************************************************************************ > GM Proprietary > > The information contained in this electronic communication and its > attachments (if any) is confidential and subject to legal privilege. > The information is intended only for use of the individuals(s) to > whom it is addressed. If you are not an intended recipient, or the > agent or employee responsible to deliver it to an intended > recipient, you are hereby notified that any use, dissemination, > distribution or copying of this communication is strictly prohibited. > If you have received this electronic communication in error, please > delete it and immediately notify me by sending a return e-mail to > the address in this e-mail. Thank you. > ************************************************************************ > > ************************************************************************ GM Proprietary The information contained in this electronic communication and its attachments (if any) is confidential and subject to legal privilege. The information is intended only for use of the individuals(s) to whom it is addressed. If you are not an intended recipient, or the agent or employee responsible to deliver it to an intended recipient, you are hereby notified that any use, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this electronic communication in error, please delete it and immediately notify me by sending a return e-mail to the address in this e-mail. Thank you. ************************************************************************ . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
