Hi, I was wondering if anyone here could provide some advice on how to do a relative valuation, this is my first ever attempt and any help would be appreciated.I have a data set of unlisted companies where I have information on book value, earnings, and naics industry. I understand that I need to find a comparable peer group, and according to the literature I have read comparable peers will have similar characteristics in terms of risk, growth, and cashflow. The universe for my peer group will be all Canadian TSX listed companies. I figure I will take that universe and break it down by NAICS. But after that I am lost as what to do. I can get the beta, growth, and cashflow variables ouf of Bloomberg for the TSX listed copmanies by industry sublist, but then what do I do? Do I come up with ranges of those 3 characteristic variables in each industry (something like manufacturing group has subgroup of high growth, med. growth, low growth) and then somehow determine/connect which range the unlisted company should fall under? If so what would be the mechanism to determine which sub-set of peers it should belong to?Or am I way off base here?

Why are you doing this? If it is a homework assignment, it sounds like you have been givenremarkedly little guidance.

First you need to determine which ratio you want to use in your relative valuation (PE ratio, Price to book, firm value multiple etc...) then you need to look at the factors that affect this multiple, for example risk, growth, payout ratio etc... Run regressions using the sector or market data that you have then plug the variables of your firm in the regressions to see if it is over or undervalued.

Hi Mballack, thanks for the reply.Given that the firms I am trying to estimate are private, I have very limited data about them, book value being the strongest measure in terms of data quality. I am trying to estimate a market value for the private firms, and was thinking that if I found a comparable set of peers in the same industry, I could somehow use their p/b ratios and apply that multiple to the book value of the private firm. I was thinking if I grouped the peers (ranges of high, medium, low, or some other grouping method) by the factors that affect the multiple (growth etc.) then I could look at the earning's growth of the private firm I am trying to estimate and decide into which grouping of peers it belongs. Then take the average of the p/b of that grouped peer group and apply it to the private firm.I have read a bit about having to run a regerssion, but have also read that central tendency measures can also be used, the simplest of which would be the arithmetic mean. And since I have a very large data set of private firms, I'm thinking the mean could be used. But I don'T know if what I described above is valid or not, or if there is some other way to apply the mean. The most difficult part of this exercise is that I can'T find any hard examples with numbers that goes through every single step, whether it be regression or some other way to do the r.v.