Nielsen salary

Originally Posted by anba

You're right, i'll be taking the job. This discussion is more of a curiosity thing.
My prior exp was with think (Booz Allen/IBM/Accenture) kinda firm. Both undergrad/masters from global Top 20 uni. The gap in resume, all i can say is.. sometimes life makes things hard for you.. for whatever reason.

I wonder if this salary offered is for the probationary period - its 6 months. Maybe change after that? As far as I know, the job is fairly advanced cause we were discussing very technical level stuff (e.g. logit/probit, hierarchical modeling, etc..) in interview.

Pls feel free to enlighten me if I'm mistaken on technical stuff described is not advanced. They said I'll be dealing with price-promotion models, marketing mix models, customer segmentation etc..

Well sometimes things are what they are so you will just have to make do with whatever pay they offer. I don't know what is Neilsen's HR policy, but as far as I know most MNCs do not increase pay after probation. If they had such a thing, I suppose HR would already have informed you when they made the offer.

I do not work for Neilsen per se, but for a vendor that provides services especially at the intelligence & market data side for them, so I have plenty of interaction with staff from Neilsen. I can't comment on your specific job, but what I can say is just like most consultancy firms, they have a tendency to inflate relatively simple stuff to make it sound real complex and academic to their clients in their marketing pitch.

At the end of the day I believe in the laws of free market. If this position is really that highly specialize and requires deep mathematical modelling skills, a global market leader like them would never stinge on that few thousand a month so that they can save cost. This kind of behavior maybe SMEs will do, but not a market leader in the industry.

Originally Posted by anba

Can I go into any in-house function at unilever or GSK or Cargill type firm. In like market research, analytics or related area.

for market research definitely yes. its so common for people to move to client-side... that is if you can survive agency-side life long enough to gain the required experience.

[anba]

Thx for the info.

Originally Posted by anba

You're right, i'll be taking the job. This discussion is more of a curiosity thing.
My prior exp was with think (Booz Allen/IBM/Accenture) kinda firm. Both undergrad/masters from global Top 20 uni. The gap in resume, all i can say is.. sometimes life makes things hard for you.. for whatever reason.

I wonder if this salary offered is for the probationary period - its 6 months. Maybe change after that? As far as I know, the job is fairly advanced cause we were discussing very technical level stuff (e.g. logit/probit, hierarchical modeling, etc..) in interview.

Pls feel free to enlighten me if I'm mistaken on technical stuff described is not advanced. They said I'll be dealing with price-promotion models, marketing mix models, customer segmentation etc..

Those are actually taught in elementary stats and econometric modules at the undergrad level... Unless u guys are talking about the theoretical properties of the models, I dun think they are very advanced.

[anba]

Fair enough.
Another question - At nielsen, my focus will probably be around the FMCG/Automobile industry doing whatever regression analysis stuff they ask me to for marketing purposes.

But, is it possible for me to switch into analytics in different industries / using it for different purpose (such as fraud, AML in financial services or join like Deloitte Analytics etc..)

Anyway, keep a look out at the public sector too. The whole public sector is jumping onto the analytics bandwagon. And if u know where to look, there are certain ministries who pays very well for ppl with ur background (Big hint: economics with strong programming background). Im toking about 70-80k p.a. for fresh grad with first class hons and ns.

The downside is that the entire service is so new to analytics that there wont be many experienced practitioners to hold ur hand (unless u r in IDA/astar). But at the same time, u r given tons of freedom to explore as wide and as deep as you want. At least that's the case for my side.

[anba]

All very interesting. Thanks for the insights!

Maybe I'm not getting this, but what exactly is so special about doing "regression analysis"? I use it on a daily basis and all it really involves is lining up variables and the associated data points followed by a few clicks on any statistical software which will provide you with all the relevant information & graphs.

I'm just a normal junior exec with a bachelor degree from Monash working in a medium size local consultancy and really don't see what is so fascinating about determining price points using regression. As long as the methodology is established (which I'm sure Neilsen is more than competent in this), it really isn't anything harder than fresh grad level accounting or financial reporting.

Neilsen's offering of 4k+ seems logical for any normal back office exec position with a few years experience, is the TS reading too much into a simple interview?

Originally Posted by Unregistered

Maybe I'm not getting this, but what exactly is so special about doing "regression analysis"? I use it on a daily basis and all it really involves is lining up variables and the associated data points followed by a few clicks on any statistical software which will provide you with all the relevant information & graphs.

I'm just a normal junior exec with a bachelor degree from Monash working in a medium size local consultancy and really don't see what is so fascinating about determining price points using regression. As long as the methodology is established (which I'm sure Neilsen is more than competent in this), it really isn't anything harder than fresh grad level accounting or financial reporting.

Neilsen's offering of 4k+ seems logical for any normal back office exec position with a few years experience, is the TS reading too much into a simple interview?

Knowing how to click on statistical software is different from knowing how to use regression analysis.
Did u do any diagnostic tests to check for normality, multicollinearity, and homoscedasticity?
if the model failed any of the test, how do u proceed?
How do u choose the the independent variables? by experience? based on empirical result? stepwise or some other forms of selection process? Is the model variation in discrete selection process a concern? If so, should I use penalty regression instead?

And that is just the model building stage. There are many more other issues to take note of at the inference stage.

Altho, none of the above may be of concern in practice in the private sector... afterall, it is jus a group of ppl who know little about statistics giving advise to another group of ppl who know even less.