Frequently Asked Questions

1. About Our Forecasting Service

2. About Our AI Technologies

3. About Getting Started In Investing

4. About the Stock Market and Trading

5. About Our Investment Performance

6. About Top-Down Market Research, LLC

7. About Subscriptions

    1.   About Our Forecasting Service

Basically, we provide forecasts on how well specific stocks and industry mutual funds will do over the coming year. Included are buy, hold and sell ratings. You may choose to follow the buy, hold and sell ratings, as we do with our own money. Or, for those of you who are more independent minded, you can use our 12-month forecasts on stocks and industry mutual funds as a supplement to your own research and investment techniques. For more information on this, read the Fund Investors, the Stock Investors, and the Keep It Simple pages.

The AI Stock Forecast is published between the 7th and 14th of each month, depending on how soon we can clean and process the raw data that is used for our artificial intelligence computer models. Immediately after we publish the AI Stock Forecast, we send an email to notify our subscribers that the latest issue is available online.

Sorry, we don't mail hard copies. If you've found us on the internet, then we assume you can access our website and print out your own hard copy. The internet is an incredible information distribution tool with which conventional mail cannot compete. It is quick, accurate, easy to use and provides a way to get time critical information into your hands as fast as possible.

No. In the past we have tested our artificial intelligence systems using weekly forecasts, but this turned out to be less profitable than making forecasts just once a month. We have found that the more often forecasts are made, the more trading you are forced to do. Trading costs create immediate losses in the form of commissions paid to your broker and also the less visible costs related to liquidity and bid/ask spreads. Limiting the forecast updates to once a month helps to minimize trading costs.

No. The AI Forecasting system focuses on long term performance and does not advocate short term trading. We know there are many advisors promoting the use of their hot lines with up to the minute stock tips. These services generate more income for the advisor and more income for your broker, but rarely more income for you.

No. Our historical testing has shown that making, say, 3 month or 6 month forecasts does not increase profits. It only promotes more frequent trading which in turn increases your costs and the commissions that you will pay to your broker.

No. The goal of the AI Stock Forecast is not to tell you when to get in and out of the market. This is called market timing and very few market gurus have been successful at attempting to beat the market this way. (For more information on this, read Other Strategies). Instead, our goal is to help you allocate your stock investments among the industries and stocks that are most likely to do well over the next year regardless of the direction of the overall market. Using this technique, our systems have been able to outperform the overall market by about 10% a year, on average.

No, we have not investigated the profitability of using our system for shorting strategies. In round numbers, the market goes up roughly 2/3 of the time and down 1/3 of the time. Also, a stock is just as likely to go up 100% as to go down 50%. So, unless one has an extremely good timing model (which we have not seen to date), using a shorting strategy can quickly make one poor. Shorting is akin to trying to win a race while sailing into the wind while everyone else has the wind to their backs. (As a side note, we have noticed that customers most often ask us about shorting strategies at market bottoms, such as in October of 2002. Anyone starting a shorting strategy at that time would have been devastated over the next year.)

In general, you should try to act on the information as quickly as possible, although it is not essential. If you have a prediction that a stock will go up in the next 12 months, it makes sense to assume that the longer you wait, the greater the chance that you will miss out on some of the profits. However, waiting a day or two to act on the information should not have a big impact on your investment returns.

The industry forecasts are made specifically for the industry groups and not the Fidelity Funds. Since there are 100 industry groups, but only 41 industry specific mutual funds, some of the funds are cross referenced to more than one industry group. When making investment decisions for the AI Fund Portfolio, we would buy the Fidelity Fund that is related to the industry that has the buy rating regardless of any other contradictory hold or sell ratings that may be applied to another related industry group. Likewise, we would not sell the fund until a sell rating is issued for the initially related industry group. This is the system that we have used to produce the investment performance shown on the AI Performance page. Take a look at the Funds / Industries cross reference page for more information.

There is no right or wrong way to do this, but it makes sense to diversify your investments in equal amounts among a number of top-rated stocks and industries. If you decide to own 20 stocks as we do in the AI Stock Portfolio, you can keep it simple by using approximately 5% of your portfolio to buy each stock. The actual percentage of each stock purchase will vary as you will probably want to buy in round numbers of shares (10's, 100's, 1000's) to make for easier trade executions. Also, it would make sense to spread the stock investments over 10 industries - 2 stocks from each industry. This will help to reduce the overall volatility in your portfolio and still allow for good returns.

    2.   About Our Artificial Intelligence Technologies

There are many people around the world attempting to apply artificial intelligence techniques (neural networks, fuzzy logic, support vector machines, genetic algorithms, etc..) to the financial markets. There have been many publicized stories of failures. Most of these failures were probably due to the misapplication of AI techniques on data that did not have a true cause-and-effect relationship with the financial markets. However, there are undoubtedly many success stories that have not been publicized.

The source codes for our AI programs have been developed from scratch and include adaptive features not found in commercially available AI software. Also, we feed our computers better data than every other service that we've found. Most, if not all, of our competitors only use historical price and volume data to train their AI systems. Although price and volume data is easy and cheap to get, the predictions that they generate do not extrapolate well into the future. This is analogous to a boy who dropped a dime that rolled under the couch, and when asked why he was looking for the coin in the middle of the room, he replied, "well, it's easier to look out here because the light is better". Likewise, just because price and volume data is easier to get doesn't mean it's the right place to look for a forecasting tool. To learn more about the data we use, read about Our Computer Models.

Yes, but we don't do that (although we have seen other people fraudulently do this). When we developed the AI models, we were very careful to avoid back testing on data that the model had been trained on. In order to properly back test a model, one must only train the model on data that was available up to the point in time that a prediction is made. For instance, when back testing and making a prediction for July of 1992, the model must be trained on historical data up to, but no later than, June of 1992. Then, for the next month's predictions for August of 1992, the model must only be trained on data up till July of 1992. This is very time consuming, but it is the right way to do it.

Each month, we update and retrain the AI models with the latest data. So, the models are continually learning as time goes on.

    3.   About Getting Started in Investing

There is a wealth of information on this web site as well as on the sites of the discount brokers and mutual fund companies. Some good discount broker sites can be found at www.tdameritrade.com, www.scottrade.com, and www.fidelity.com.

It's simple. Just contact them and they'll send you all the information and paperwork you'll need to open up an account. For stock investing, look for a good discount broker such as TD Ameritrade, Scottrade or E*Trade. These will only charge you about $10.00 per online trade. For mutual fund investing, if you're using our industry group forecasts, there is only one good choice - Fidelity. They have the best selection of industry specific mutual funds. The industry funds are called the Fidelity Select Portfolio funds and include a total of 41 mutual funds that are industry specific. Instead of using a broker to purchase the funds, you can avoid brokerage commissions by opening up a mutual fund account directly with Fidelity. With a Fidelity account, you can trade among the Fidelity Select Portfolio funds for free. For a list of the Fidelity Select Portfolio funds, go to the Funds / Industries page.

Yes. Stock brokerage firms as well as mutual fund firms can open an IRA account for you. Also, if you are leaving your current employer, you can roll over your 401K money into an IRA account without being taxed. You may want to discuss this with a personal financial advisor - but be forewarned that the advisor may want to sell you a mutual fund that pays him a sizable commission.

Well, that depends. The stock market can be very volatile. It offers great potential for gains, but also, on occasion, great potential for losses. The younger you are, the more risk you should be willing to take since you have more time until retirement to recoup any losses that you may incur. A simple formula used by many financial advisors states that the percent of your investment assets that should be invested in the stock market should be equal to 110 minus your age. This formula should be adapted to fit your personal financial situation and your own risk comfort level. You may want to discuss this with a personal financial advisor.

Fidelity has a minimum requirement of $3,000 to invest in any of their Fidelity Select Portfolio mutual funds. So, $3,000 is the minimum that you need. If you plan to invest in the same 10 mutual funds that are published each month in our AI Fund Portfolio, you will need a minimum of $30,000 to get started. But, if you don't have $30,000 to start with, you can begin by investing in a few funds and then increase the number of funds over time as your investment portfolio grows. Generally, the more funds you invest in, the less volatile your overall portfolio will be.

There is no right or wrong way to start out. If you are bold and are anxious to get started, you can buy all 20 stocks in the AI Stock Portfolio at once. If you are more risk-averse, you can ease in slowly, just buying the few stocks with "buy" ratings that are added to the AI Stock Portfolio each month, leaving the rest of your investment dollars in cash. The advantage of investing in all 20 positions at once is that if the portfolio does well over the next year, you will have participated in the profits fully. The disadvantage would be if the portfolio does poorly over the next year, then consequently, you would have been better off taking a more cautious approach. If you decide to ease in slowly by just investing in the new positions that have "buy" ratings, it may take a year or two to get fully invested. The "AI Performance" and "Stock Trades" pages on this web site will give you an idea of the kind of volatility and profits that you can expect on a month-to-month basis if you are fully invested. This information is what you need to consider when deciding how to start out.
 

No, you don't have to. If you are only investing a portion of your savings into the stock market and you have other investments that provide diversification, then you probably don't need to buy 20 stocks or 10 mutual funds. But you should understand that the portion of your money that is invested in the stock market will tend to be less volatile the more stocks you own. For instance, during 1999, in the AI Stock Portfolio, the largest single stock profit was 138.9% and the largest single stock loss was 62.3%. A single stock investment can be quite volatile. During the same period, the AI Stock Portfolio, which held 20 stocks at a time (including the two just mentioned) had a 27.0% profit, but experienced a maximum pull back of only 15.1% during the course of the year (see the AI Performance page). Diversification helps to limit volatility. You may want to discuss this with a personal financial advisor.

    4.   About the Stock Market and Trading

Market orders are easy to use and always get filled immediately at the current price that someone else is bidding or asking. However, if the stock you are attempting to buy or sell is thinly traded (not many shares are traded each day), using a market order can be dangerous. For example, if you are attempting to buy a stock on a particular day using a market order and only one person is attempting to sell but is asking a very high price, you will automatically be filled at the high price that the seller is asking. So, if you are going to use a market order, it is prudent to first check on the current bid and asking prices on your discount broker's web site before placing the order. In general, market orders work well for stocks that trade more than 100,000 shares a day. Limit orders should be used for stocks that are thinly traded. With a limit order, you can specify the maximum price that you are willing to pay, or the minimum price that you are willing to accept for a stock. However, with a limit order, your trade may not be filled if the stock does not trade at your specified limit price.

No, not if you're following the AI Stock Forecast system. The system focuses on long term performance and is based on the principle that to be a successful investor, you need to be patient. All stocks experience periods of distress. Sometimes these periods are the best buying opportunities, sometimes not. How do you know which is which? You don't, but following the AI Stock Forecast system helps you to make a rational and unemotional decision on when to sell a stock. To offset the effects of the losers that you will inevitably have in your portfolio, you need to be diversified. In a diversified portfolio, you'll have some big winners and some big losers. Most studies indicate that you need to have at least 8 to 10 stocks in your portfolio to smooth out the volatility.

Lastly, many chart based (technical) traders use stop loss strategies and their stop loss orders tend to accumulate at or near the same prices. Other powerful traders know this and will often push a stock past these points just to "take out" the stops. Why? Because they know that around these points that there is free demand to offset their orders. In other words, these traders will sell a stock heavily to push the price down to where they know they can buy it back cheap as the sell stops are hit.

Yes, you could do this. However, there are less industry specific options with ETF's than with the Fidelity Select Funds. At the time of this writing, there are only about 20 ETF's versus 41 Fidelity Selects that are industry specific. For help with this, look at our Funds / Industries cross reference page.

Before turning to AI techniques, a lot of our research was focused on historical testing of technical trading systems. Despite years of work, we were not able to find any benefit in using technical analysis for stocks. It is tempting for investors to flip through a stack of stock charts looking for price and volume patterns, but the human mind is often easily fooled into seeing patterns that do not exist. In order to really test a technical trading system, one needs to use a computer program that can objectively test a trading methodology over thousands of stocks and over many time frames, making sure to apply the proper trading costs. After doing this, we did not find any reliable technical indicators that are profitable over the long term.

The whole validity of technical stock analysis hinges on one question - is it reasonable to think that the future price of a stock can be predicted by looking at past prices? In our opinion, this makes about as much sense as turning the seat around in your car, mounting the steering wheel on the back seat and trying to drive forward while looking out the back window!

To date, we have not seen this phenomenon. Unlike many other investment newsletters that only recommend a handful of stocks, the AI Stock Forecast posts ratings on almost 5000 stocks. Approximately 700 to 1000 stocks have buy ratings at any one time. So it is unlikely that this will ever be a problem for our subscribers.

    5.   About Our Investment Performance

We maintain two model portfolios that are updated each month using the buy, hold and sell ratings issued in the AI Stock Forecast. The AI Stock Portfolio invests in 20 stocks and the AI Fund Portfolio invests in 10 mutual funds. The portfolios remain 100% invested at all times. The updated performance of these model portfolios can be viewed each month on our AI Performance page.

For the purpose of computing our investment performance, on any month in which any positions are sold, all of the proceeds from the sales are equally invested into the same number of new positions. For instance, if three stocks are sold for a total of $7500, then they are replaced by three new stocks that are bought for $2500 each.

The S&P 500 Stock Index is a composite of 500 stocks in which every industry is represented. It is a very good yardstick to use if you want to measure how well you are doing compared to the "average" stock.

The prices used for stock and fund trades are taken on the close of the publication date. The trade dates and prices are listed in the tables on the AI Stock Portfolio and AI Fund Portfolio pages. The AI Stock Forecast is normally published before the next trading day after the publication date. Also, subscribers will typically receive an email notification before the next trading day after the publication date.

For yearly performance data, instead of computing from January 1st to December 31st, we compute the annual returns using the closing prices on the January publication dates. So, for instance, in computing the return for 2004, we used the closing prices on the publication dates of 1/10/04 and 1/14/05.

    6.   About Top-Down Market Research, LLC

No, unlike many other investment professionals, we do not receive any payments for recommending any stocks, mutual funds, brokers, or mutual fund companies. Our unbiased research is completely independent and has no conflicts of interest. Our service is provided solely for the benefit of subscribers to help them take control of their own investments.

To empower investors to take control of their own investments. Our service helps provide the individual investor with another option as opposed to turning over their hard-earned money to a highly compensated financial "professional". Using our forecasting service and a low cost discount broker, the individual investor can beat the "professionals" at their own game.

Yes. Because we believe in our technology, the principal owners of Top-Down Market Research, LLC invest in the stocks and funds that make up our model portfolios, the AI Stock Portfolio and the AI Fund Portfolio.

No, we wait until the next trading day after we publish to make our trades. This policy maintains our integrity and ensures that we cannot be accused of "front-running" our customers. (Front-running is the unethical practice where a financial professional knowingly buys or sells for his own account before information is made available to a client or the public, thereby profiting from the influence on the price made by the customers order.)

No. Currently, we are not a registered investment advisor, so we are not authorized to provide personal financial advice to clients or to manage a customer's accounts. Instead, our goal is to provide information that empowers our subscribers to take control of their own investments and avoid the excessive management fees levied by many financial professionals.

    7.   About Subscriptions

No. We will not automatically renew your subscription. To renew, you will need to re-subscribe again on this web site.

We will send an email to you on the day that your subscription expires. The email will include a link to our subscription page so that you can easily re-subscribe, if you want to.