Stock valuations

Valuation results generated by your model show drastic under- or overvaluation for some stocks. How confident are you in your valuation methodology?

It is all about input parameters to the valuation model. Our valuation algorithm uses historical (i.e. backward-looking) data. Stock valuation, however, is all about the future. We do not pretend to have a crystal ball, and, instead, explicitly assume that the past data is the best predictor of the company's future performance. There are many drawbacks to this approach that, in our opinion, are far outweighed by its major advantage: all stocks are valued on the same objective basis, without interference of any subjective judgement. Having the same 'playing field' for valuation of all stocks allows for fair objective comparisons. Nonetheless, investors are strongly advised to verify all model inputs and adjust them according to their view of the company's future performance.

Your stock valuations are very sensitive to even slight changes to the model input parameters. How do you determine what numeric values of inputs to use?

We employ a rather mechanical approach to determining numeric values of inputs. They are determined by a set of calculations performed on historical financial statements of the company in question. All valuation models are very sensitive to input parameters, which should correspond to anticipated future parameters. Future performance, however, does not necessarily has to follow historical data that we use. Therefore, an extreme care should be taken by the investor in selecting values that will best match the company's future performance.

I have been using a DCF calculator on another site that defaults to the discount rate of 11%. I do not understand why your valuation model starts with various discounts rates many of which are below the average return of the market over time. Would you be so kind as to explain this?

That's one of a distinguishing features of our valuation model. In calculation of the discount rate we use a calculated probability of the company's default in the period of one year from now (which is usually quite low) and then escalate that discount rate in subsequent years. Site users can change both the initial discount rate and the escalation factor (discount rate multiplier).

Why and how from one week to another a company/stock can go from being substantially undervalued to substantially overvalued?

This could happen primarily when new financials become available that change the model's input parameters. We use a computer model for valuation - therefore the input data is so important. Besides the company financials, the valuation is also affected by changes in macroeconomic parameters. In addition to the above, from time to time (quire rarely) we introduce changes to the valuation model itself based on our understanding of the market fundamentals - to make it more theoretically sound.

Why valuation result is not available for some stocks?

This primarily has to do with availability and completeness of at least two years of financial data. For some small companies (especially the ones in the "penny stocks" category) this could be problematic. Another reason: the financial data we have on a particular company is not consistent with the valuation algorithm we use, which puts in question the validity of the valuation result.

Market valuations

The distribution of X-FIN ratings for every stock index and sector shows that the majority of stocks is rated as "sell" even for indices and sectors that are fairly priced or undervalued. How is this possible?

The distribution of ratings alone does not show the full picture. Theoretically, on the downside the price of a stock could only go to zero (the loss of 100%), but on the upside it is unlimited. So, a handful of stocks can drive a stock index higher. Therefore, it is OK for the majority of stocks to be rated as "sell" even in a fairly-priced or undervalued market.

Performance

Do you have any information for years prior to 2015 about your valuation model's performance?

Not really. We used the model for selecting stocks for our own portfolio and we monitored their performance, of course - but not a wider universe of stocks. Besides, through the years the model underwent a few major revisions with the latest one done in 2014.

How would you recommend using your valuation model: buying stocks from the "strong buy" list at the beginning of the year and then selling them at the end of the year or doing a more frequent rebalancing?

We recommend establishing long-term positions - for at least a year. Our stock valuations are updated regularly throughout the year and you may act on them whenever you want. It is just for comparison purposes that we track the full-year performance.
Of course, there is no guarantee that you will make money following our model's recommendations, but if your portfolio is well diversified the odds are that you will beat the market. You would still need to make your own research on the stocks you select - as our valuation model could be viewed only as a fancy stock screening tool.
And one more thing: don't strive for the maximum gain, but rather for the maximum certainty of a gain.

You need to know that your model for valuation need some work! I had put faith in it and unfortunately lost \$XXX. I am referring to Banco Santander (STD then SAN). The stock has been going nowhere but down despite your assessment of it being way undervalued more than 4 years ago....

First of all, the model is not to be applied to valuation of banks and other financial companies. Banks' market value is a bellwether of the state of the economy in a country. We are not aware of any model that values banks with any respected level of accuracy. If you plan to invest in a bank, you should first project a macroeconomic situation in the country, then analyze composition and quality of the bank's portfolio, and, last but not least, see if the bank's management are smart and responsible people.
Second, contrary to your experience, there is strong statistical evidence showing that the model works. The key word here is "statistics", which means that the portfolio should be well-diversified. If you put all of your money in one stock, your could expect either to gain a lot or to loose a lot.
Also, the model is just a tool, with input parameters taken from historical financial statements. The investor should adjust these inputs according to his view of the company's future performance.
And the last observation: once you've invested, you should monitor the company/bank you invested in. Over a very long period of time many companies go out of business - this is natural. The best performing company today could be among the worst-performing ones in several years. We see this happening all the time.

S&P 500 Validation

What is the relationship between the intrinsic value of the S&P 500 Index and the S&P 500 validation index?

Intrinsic value of an index is our model's estimation where the index should be if it is calculated with intrinsic values of constituent stocks instead of their market prices. The S&P 500 price movement validation index, on the other hand, just confirms (or otherwise) price movement of the index - its absolute value is irrelevant (the base for it was chosen arbitrary). So, these are totally different animals.

How often information on the site is updated?

Company financial information and stocks valuations are updated every month; stock prices are updated every day.

Subscriptions

I was wondering if you could offer 6 month and quarterly subscription and monthly packages as well.

At the moment we do not have plans to offer other packages. Most of the material on the site is available for free - though with some restrictions on usage, but still quite adequate for most individual investors.