Of late I have started self-learning. I have bought a few well-advised Statistics books such as Statistics for Management by Levin and First Course on Probability by Ross. I observe that Ross' book has much more coverage than that of Levin's. I think Ross covers probability from Mathematical Statistics point of view rather than that of Business Statistics. Levin's does not have as much coverage on probability as Ross (In my humble opinion). And there I get confused.

Given that I want pursue a career in Data Analysis, should I follow Ross book along with Levin's? Will Levin's book's coverage of Probability will suffice? Do you think following Ross' book First Course in Probability would be difficult and I should first start with Levi?

Could you please give your guidance.

Warm Regards Sabya

  • $\begingroup$ What kind of data analysis...if you're going into mathematical finance kind of analysis, you want Ross' book for sure $\endgroup$ – user76844 Feb 10 '14 at 19:02
  • $\begingroup$ My two cents: start on Ross, do the exercises, then get a good statistics book (Wackerly is what I used when I was learning stats, but Wackerly also covers probability). Then grab a measure-theoretic probability text if you have a good analysis background. $\endgroup$ – Clarinetist Feb 11 '14 at 17:18
  • $\begingroup$ Thanks a Clarinetist. Your suggestion will prove very useful novice like me in learning Statistics and Probability. $\endgroup$ – Sabyasachi Mitra Feb 13 '14 at 17:48

I do data analysis for a living, but I'm not a mathematician or statistician by training, so my answer might be biased. Furthermore, as Eupraxis1981 mentions, it will depend on what kind of analysis you want to do.

None the less, I would say that as an analyst working with data on education, I never resort to probability theory. The same is true for my collegues in marketing and a lot of business. I think you would be much more well-advised to get a firm grip on programming.

However, obvious caveats apply: if you want to go into finance, you're gonna need to know the definition of a stochastic process and how to apply that. But always remember that statistics is, in it's heart, mathematical engineering and here, as in other parts of engineerig, ingenuity and creativity is of the first importance.

  • 1
    $\begingroup$ At the same time, statistics is so badly and so routinely abused in its applications that one can only hope that more practitioners would be better taught in its principles. $\endgroup$ – Did Feb 11 '14 at 7:03
  • $\begingroup$ Thanks both of you for the suggestions. I am a Healthcare Analyst and learning Stat to work in areas of Risk asessment & risk adjustment, fraud detection, claim leakage, cost forecasting, etc. There are various models present in the market such as Clinical Risk group, diagnosis cost groups, episode risk groups. To understand these models & do analysis using these models I should have strong statistical foundation. Also, I have an inclination towards analyzing economic data (not finance) and econometrics. Given this is the requirement can you please tell me what should I more emphasize on? $\endgroup$ – Sabyasachi Mitra Feb 11 '14 at 7:07
  • $\begingroup$ @Did, I'm not advocating the abolition of probability theory! What I am saying is that for a lot of the work that you do as an analyst, the opportunity cost of getting a good grip on probability theory over learning, say, SQL or another practical tool would be very high. Sabya, I'd recommend getting a good guide to econometrics, Wooldridge or Greene maybe. Angrist and Pischke's "Mostly Harmless Econometrics" is a slightly easier option. $\endgroup$ – abaumann Feb 11 '14 at 9:31
  • $\begingroup$ Hello Friends - Thanks for your time and advice and sorry for my delayed response. I think I got quite a few useful suggestions. $\endgroup$ – Sabyasachi Mitra Feb 13 '14 at 17:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.