Also, one thing, if it is a new sector, consider if you can do some paper trading based on actual returns in the past. For instance, get annual reports from se

Also, one thing, if it is a new sector, consider if you can do some paper trading based on actual returns in the past. For instance, get annual reports from several companies from 2004. Try to get a survey of companies in the sector, and not be influenced based on what you know now about the companys. If you're married it would be best to have your spouse anonymize the companies by copying down the key figures from the annual reports, and then the price on Jan 30th, for each of the next 5 years. Have her keep the company key seperate. Analyze them on that basis, then look into the prices on the stock market in January 2005. Use those prices and discounted cash flow analysis to predict what their prices will be in 2010. Invest a fake $150,000 in the companies based on those calculations. Then wait a week. Each week, look up the prices for the companeis a year later and decide if your thesis still stands. For this example, lets just pretend like you have no access to news, no annual reports for 2005, just the price on Jan 30th, 2006. Decide to re-allocate if you want, but pay $100 for each transaction (sales and buys). IF you're down and you sell then you're locking in that loss. Do this for the next 5 weeks until you've covered the time period to Jan 30th 2010.... and then look to see how your companies did. Try to draw your conclusions before you connect the fake companies names with the real companies. Don't forget to account for dividends in the interim as well. This can be very instructive-- but be sure to learn about how DCF and current trading price affect risk. IF one of the companies did poorly, was it because of bad management? An externality that was unpredictable? etc.