Efficient markets, EPV and Graham valuation methods

Saturday, Jan 26 2013 by

Here are some graphs: they show a bunch of companies placed according to the automatically generated Earnings-Power (vertical-axis) and Graham valuations (horizontal) from this site. EPV is supposed to consider the firm as it stands today, Graham as the firm it will grow into. AIUI. Overvalued is top left, Undervalued bottom right. No particular rationale for the choices of firms: some are people's picks for 2013, a few others chosen for quality, flakiness etc.

Now, surely some quantitative factor from the balance sheet will correspond to the valuations. Companies are wrongly valued for good reasons,right? So I've also culled all the other data relating to all these firms, so I can use some bit to colour in the tickers. This first one uses the Piotroski score [hope it fits the page ok..]:

And this one the ROE..

And so on. I'll be brief because I've a headache coming on from programming this: there's no bloody pattern! Not even the last one. The valuation methods Stockopedia provide seem really sensible - so why is there no bunching of quality in the overvaluation sector.. ? Please discuss.

I also do painting and decorating. :-Q


Unlock this article instantly by logging into your account

Don’t have an account? Register for free and we’ll get out your way


As per our Terms of Use, Stockopedia is a financial news & data site, discussion forum and content aggregator. Our site should be used for educational & informational purposes only. We do not provide investment advice, recommendations or views as to whether an investment or strategy is suited to the investment needs of a specific individual. You should make your own decisions and seek independent professional advice before doing so. The author may own shares in any companies discussed, all opinions are his/her own & are general/impersonal. Remember: Shares can go down as well as up. Past performance is not a guide to future performance & investors may not get back the amount invested.

Do you like this Post?
7 thumbs up
0 thumbs down
Share this post with friends

6 Comments on this Article show/hide all

snickers 4th Feb '13 1 of 6

I'll have a second bite at this subject, with improved graphics. A second chance for Stockopedia readers to reply, anyway.

The point of this graph is to detect overpriced-poor and underpriced-quality firms. Firms are displayed here as dots whose size, redness and opacity are controlled by 3 financial measures, and in this first snapshot big, red and opaque is bad.

So, I'm looking for bad firms in the top left to sell: GPOR, Salamander, Shanks, Alumasc, Melrose; and bad ones on the other side to not chase: Dart and Skywest (air lines), Creston, JDSports. I had a look at Skywest: there's a takeout offer been made at the current share price, so the parties involved surely aren't using either of the valuation measures graphed here.

On the other hand, making big and red into positives I get this:

So Zytronic, Haldane look overvalued for good reason, and ACHL(chinese oranges) HSX and CGL(insurance), and something called SRG (escaping off the edge) look interesting. If you cast an eye over the names, what's your opinion of the distribution? Does the market disagree with the financial numbers for good reason?

On the third other hand, using a different set of financial measures produces this unhelpful picture:


| Link | Share | 1 reply
Edward Croft 4th Feb '13 2 of 6

In reply to snickers, post #1

Snickers - it would be great if you could add some axis labels to some of these charts as I have to admit I'm not sure in the main post which chart relates to which idea!

Blog: Follow @edcroft on Twitter
| Link | Share
dangersimpson 4th Feb '13 3 of 6


It would be also good to explain what you are trying to do with these charts, what is your thesis, what are you looking for and why, how are you going to test this thesis/results?

I think the reason you are not getting any feedback is that no one understands what you are trying to show or investigate so its impossible to have an opinion. It's like the results of a school science experiment but without any details of what the experiment was!

Also four factors on one graph is almost impossible to get my head round (unless the fourth is time and the graph is dynamic) I suggest you stick to 2 or at most 3 and increase the number of graphs.

| Link | Share
snickers 4th Feb '13 4 of 6

Oh my god. And I,m typing this reply on a tablet....
First off, there's an image placeholder at the top of my response to my own post. I can't remove it now there have been further comments..
Aside from that, what I'm interested in is the apparent discrepancy between these 2 company valuation methods, (which seem, as I said at the top, to be reasonable, logical etc...), and the market's opinion of the companies. Especially the EPv method.
The graphs, ok, they' re a bit of a mess. I have artistic leanings... & Yes, school science project is what it is.
Companies are placed according to the 2 valuation methods, graham and earnings-power. Objectively overvalued is top left, undervalued bottom right. Size etc is according to other balance sheet items. All data from this site.
But I'm trying to show that , actually, the Data is a mess, and there's not a lot of logic here. One indicator points one way, another points another . But I expected that there would be some overall pattern among many companies relating valuation to some unequivocal financial measure. Is the market logical? The general opinion is not. Is the EPv method logical? Are neither ? I suppose I'm asking in part because I'm paying a fee for this.

| Link | Share
dangersimpson 5th Feb '13 5 of 6

I guess what you are showing here is that the EPV & Graham methods are not particularly correlated in the values they give. I don't think that is particularly news since you'd expect different methods to give different values. It's the same when you include P/E or EV/EBITDA since these are really just other short cuts for a DCF using steady state conditions but a different method.

If they were perfectly correlated that would only tell you that the measures agree not that the value they generate is anywhere close to fair value or even that 'undervalued' stocks will outperform 'overvalued' - for example the measures may just be wrong or the share prices may only correct over a time period that we couldn't trade such as milliseconds or decades (unlikely I know but that's the reason that people test great theories and sometimes the theory doesn't work in practice e.g. against the EMH theory low beta stocks outperform high beta.)

So what would be more interesting is whether companies that have a high upside to Graham and EPV valuation outperform those with low or negative, and whether adding Piotroiski or ROCE improves this strategy. The best way to do this would be with the existing screener tool and a couple of model portfolios. That way you have a much wider dataset and some idea whether these measures are useful beyond the theory.

I'd look forward to seeing some results if you decide to follow up on this.

| Link | Share
snickers 5th Feb '13 6 of 6

Thanks,yes. I am going to watch how the outliers on my plot perform. Shame there aren,t more of them though. And also your point about p/e as an equivalent for dcf method: using that as the dot-colouring method does produce a nice (but not useful) pattern.
Hurrying to work..

| Link | Share

What's your view on this article? Log In to Comment Now

You can track all @StockoChat comments via Twitter

About snickers


contact cxhris at mail.com

Stock Picking Tutorial Centre

Let’s get you setup so you get the most out of our service
Done, Let's add some stocks
Brilliant - You've created a folio! Now let's add some stocks to it.

  • Apple (AAPL)

  • Shell (RDSA)

  • Twitter (TWTR)

  • Volkswagon AG (VOK)

  • McDonalds (MCD)

  • Vodafone (VOD)

  • Barratt Homes (BDEV)

  • Microsoft (MSFT)

  • Tesco (TSCO)
Save and show me my analysis