The following is a list of indicators you may see in either our data files or historical html files. In the case of historical data, of course, indicators refer to the data at the time of measurement, not data from the most current market session.

At any time, there may be a few indicators in our data that are not found below: just ignore these.  Most of our proprietary indicators involve stock/stock correlations...you can find a tad of discussion on this topic at our "seasonality and market trends spiel" page.

1. leadership : this indicator attempts to quantify whether a particular stock is a market leader or not.  The algorithm is a bit complex, but you can get a rough description of it here.  The indicator was added in March of 2004, so you won't find it in older data tables.
2. high-close : the percentage difference between the high and close of the stock.
3. close-low : the percentage difference between the closing value and low of the stock.
4. open-yclose : the percentage difference between the opening value of the stock and the prior day's close.
5. industry : the number we assign to the stock's industry group.  For more information on these numbers and industry groups, click here.
6. recent price : the closing price of the stock in question
7. stdev : the 60 day volatility of the stock in question. Our computer automatically throws out stocks with excessively high volatility, since it's possible that we've missed a split or large dividend that would unfairly bias our data. For the statistically inclined, we used to include skewness and kurtosis indicators, but they rarely seemed to be of use as predictors.
8. corr : we identify the stock that best matches the stock in question (out of about 2000 stocks), and output the "r" value that you'd get by using linear regression.  Such an indicator would be of interest to those interested in "pair trading".  A high value indicates that a highly correlated stock was identified.  A low value indicates that the stock tends to trade more as a "free agent".  This indicator was added in late 2004...it'll be percolating through our data tables over the next year.
9. pair_dif:  we output the % difference between those two (strongly correlated) stocks' performances over one day.  We subtract the gain of the correlated stock from that of the stock in question, so pair traders would generally be interested in large negative numbers if they intend to go long on the stock in question.  The indicator was added late 2004.
10. pair_dif20:  As above, but here we're looking at the 20 day % difference between a stock and its best "trading pair".
11. prd_lastyear : the percentage gain in the period to be measured, last year.  If, for example, we're trying to predict market behavior for the month of October, we'll look at the stock in question's performance last October.  Obviously, the indicator becomes rather irrelevant when looking at very short time periods (e.g. one day), since we can hardly expect the one day performance of a stock 252 days ago to be a strong predictor.  The indicator was added in September 2003, and won't be found in prior data.
12. prd_3yrs : the compounded gain of the stock in the same period over the last 3 years.  If a certain stock gains 20%, 30%, and 10% in successive Junes, for example, you'll see 1.716 (1.2*1.3*1.1).  There's an issue with this indicator that is not present with other indicators: it requires at least 3 years of data prior to the period being analyzed.  Many stocks don't have this history, so you'll only see this indicator in situations where it makes special sense to include it (i.e. it creates problems when you mix stocks that do have this history with stocks that don't).  In cases where we do mix the two types of stocks, a -1 signifies the absence of the necessary historical data.  In some cases, this -1 might be instructive in and of itself: for example, it's possible that stocks that lack a long term history might be precisely the ones that make significant gains or losses over a certain period of time.  This indicator was added in September 2003.
13. prd_max : the compounded gain of a stock over the period in question, using all the historical data we have available.  If you see .50, that means a 50% gain would have been gotten by buying the particular stock in the period of interest (over the period for which we have historical data on the stock), and selling at the end of the period.
14. yeargain : the percentage increase or decrease over a period of 252 days.  If you want to know a stock's relative strength in one of our downloadable files, just sort the file via this indicator and see which stocks have outperformed.
15. 3monthgain : the percentage increase or decrease over a period of 62 days. The data is annualized, so it's possible that you'd see a loss of greater than 100% (e.g. a 30% loss over three months is a 120% annualized loss).
16. 1monthgain: as with 8), but the data is measured over 20 days.
17. gain1: the stock's one day percentage gain or loss (not annualized).
18. gain2: as with 17), but the gain or loss from the open to the close two days ago.
19. gain3: as with 17), but the gain or loss from the open to the close three days ago.
20. gain4 : as with 18), but the gain or loss from the open to the close four days ago.
21. momentum : here, we've set up a little formula that awards high points for stocks that have been on an upward roll over the last week, and negative points for losers. We don't expect this rating to be the end-all and be-all for momentum traders, but perhaps it can be an aid in the screening process.
22. afterhours_stat:  this indicator looks at the last 30 trading sessions of the stock in question. If a stock has had a strong tendency to gain in the afterhours, the "afterhours_stat" will be a high number. To avoid the possibility that a single large gain or loss would skew the indicator, we don't allow gains or losses of greater than 5% to be included (i.e. a 17% gain in one session would be entered as a 5% gain).
23. high_moment:  if the stock has had a strong tendency to close near its high over the last 5 days, you'll see a high number.
24. open_moment:  if the stock has had a strong tendency to open above the previous day's close, you'll see a high number.
25. resistance1: you may see this indicator in data generated after 8/2003. We look at the last year of stock data, and try to find the price range at which the stock has the greatest tendency to get stuck. Then we calculate how far the current price is from this value.
26. resistance2: as with #25, but this time we look at the second stickiest price.
27. resistance3: as with #25, but this time we look at the third stickiest price.
28. inertia: you may see this indicator in data generated after 8/2003. It's a rough measure we've devised to measure the tendency of a stock to get stuck on resistance points. A high value indicates a strong tendency to plateau.  Technically-oriented folks should enjoy searching via this indicator...the chart of a stock with a high value has a very different character than that of a stock with a low "inertia" value.
29. ind_1day: the percentage gains of the industry group in question over the last day.  This indicator, and the next four, were added in January 2004.  We'll include these indicators in situations where we compound historical data with the intention of ascertaining the extent to which buying stocks in strong industry groups is a good strategy.
30. ind_1week: as above, measured over 1 week.
31. ind_1month: as above, measured over 1 moth.
32. ind_3months: as above, measured over 3 months.
33. ind_1year: as above, measured over 1 year.
34. volume1_3: yesterday's volume divided by the average volume of the last three days (yesterday's volume inclusive).
35. volume3_10: the three day average volume divided by the ten day average volume.
36. volume10_60: the ten day average volume divided by the 60 day average volume.
37. markcap_est: without digging around for the capitalization data of every company we analyze, we make a rough estimate of it by multiplying the stock value by the average volume over the last 3 months: companies like Cisco Systems and Intel will have high values, and lesser known companies lower values.
38. best_p: proprietary.  We'll be phasing this particular indicator out in the future, but you may find it in our older data tables.
39. pslice97: proprietary. Do keep an eye on this and the following seven indicators, as they often beat out more commonplace indicators in terms of forecasting stock moves.
40. pslice47: proprietary.  We added this indicator in 8/03, so you won't find it in older data.
41. pslice3: proprietary
42. vslice97: proprietary
43. vslice47: proprietary.  We added this indicator in 8/03, so you won't find it in older data.
44. vslice3: proprietary
45. big_pslice: proprietary
46. big_vslice: proprietary
47. pave: proprietary
48. pstdev: proprietary, though the end result is often quite similar to the common standard deviation measurement.
49. pskew: proprietary
50. pkurt: proprietary
51. vave: proprietary
52. vstdev: proprietary, though the end result often parallels the standard deviation measurement.
53. vskew: proprietary
54. vkurt: proprietary
55. mov_ave5: the five day moving average
56. mov_ave20: the twenty day moving average
57. mov_ave100: the 100 day moving average
58. break_ave: the percentage difference between the 100 and 20 day averages: a positive number indicates that the 20 day average exceeds the 100 day average. Some technical folks swear by this one, though we don't see it popping into our best indicator columns more frequently than a lot of other indicators.
59. yest_market: the gain or loss of the Russell 3000 over the previous period.  If, for example, we're interesting in predicting gains over the next 20 days (about one month), this figure will be the performance of this index over the prior 20 days.  Obviously, this indicator could have been more aptly named, since we're often not looking at the percentage change in just yesterday's ("yest") market.
60. mark_result: if we're looking at historical data, this will be the gain or loss of the Russell 3000 in the next (future) market session.  Obviously, we can't use this as an indicator of future performance.
61. ups/downs: the ratio of gaining stocks versus losing stocks in the previous market session.  We usually omit this indicator when crunching data.
62. prcvd_risk: here we divide the volatility by the stock price.  The idea is simply that cheap stocks are often regarded as risky, whether or not the volatility justifies that label.  A high value means that the stock is both volatile and cheap.
63. recent_std : the volatility of the stock over just the last 10 days.
64. ptotal: proprietary
65. pftotal: proprietary.  It's strongly multicollinear (i.e. almost the same) with our pave indicator, so we eliminated it after 8/03.  If you wish to use this indicator in making investments, just substitute the pave indicator.
66. t_pratio: proprietary
67. t_vratio: proprietary
68. ptot_neg: proprietary
69. stock_ind: proprietary
70. vtotal: proprietary
71. vftotal: proprietary.  It's strongly multicollinear with our vave indicator, so you won't find it in data generated after 8/03.
72. vups/vdowns: the ratio of stocks whose 1 day volume exceeds the average volume over the last five days versus stocks whose 1 day volume is less than the average volume over the last few days.  When crunching data, we normally exclude these numbers.
73. vstock_ind: proprietary
74. fstdev: the future 60 day volatility of the stock in question.  Obviously, this only applies to historic data older than 60 days.  Otherwise, the column is left blank.
75. annualized gain: For our most current data file, this column will be left blank.  Normally, we offer historic data in html summary form, but if, for some reason (e.g. a subscriber request), we present historic data in this form, you should know that this will be the stock's gain or loss, projected over one year. If, say, a stock gains 1% in the course of 1 day, you'll see 252 in this column (we assume 252 trading days in a year). If we are looking at a 1 month trading period (about 21 days), we would multiply the gain by 252/21. This may seem a bit convoluted, but we find it helpful in comparing the merits of short-term, mid-term, and long-term strategies. Given a choice between a day trading strategy and a longer-term strategy that produce similar annualized gains, we'd recommend the longer strategy because our annualized gains do not factor the effect of commissions.
76. adj_gain: the annualized gain (see just above) divided by the standard deviation of the stock in question.  This is the risk adjusted gain of the stock.  Again, for non-historic data, this column is blank.

The following fundamental indicators may appear in html summary tables generated after Oct 31, 2003.  They are NOT found in our data files: we're forbidden to rebroadcast these indicators.  This should not be a problem since there are many sites that allow you to screen stocks via a plethora of fundamental, as well as technical, indicators.  If our html tables show that stocks with a high dividend should do well in the future, just visit one of these sites and screen for these sorts of stocks.

1. BookValue/Price: the company's book value divided by the current stock price.  Often, you'll see price divided by book value, but here we're simply trying to avoid divide by zero errors in the case of a zero book value.  If the value is high, then the stock is expensive in relation to book value.  If it's low, then it's cheap. 
2. CurrentRatio: current assets divided by current liabilities
3. EPS_percent: the % change in earnings per share from last year's quarter to this year's same quarter.
4. PE_ratio: the price/earnings ratio.  Extra items are subtracted out.  There's a bit of a problem in sorting stocks via PE_ratios: normally, a low PE ratio is considered something like 5.0.  Such a company is making a lot of money relative to its share price.  If the PE is -5.0, the company is losing money.  Yet as far as the computer is concerned, the -5.0 PE is lower than the 5.0 PE. And a PE of 500 means the company is just barely making a profit. Sorting these PE's from top to bottom would be unrevealing..from top to bottom you have:  barely profitable-> profitable-> losing a lot of money-> losing a little money. Therefore, we take the inverse of the PE.  A company with a PE of, say, 6, will get a high numerical ranking.  A company that's just barely making money would be found in the middle of the pack.  A company losing minor sums of money would be below that, and those companies that are losing scores of money would get a ranking of 1. 
5. Institutions: the % of the company that is owned by institutions
6. ProfitMargin: income/revenue, as a percent.
7. QuickRatio: a stricter measure of ability to pay current debts than CurrentRatio
8. Dividend: expressed as a percent.
9. Debt/Equity: self-explanatory
10. CashPerShare/Price: If the number is relatively high, it means the company has a lot of cash on hand.
11. NumInstitutions/Marketcap: we divide the number of institutions that hold the stocks by our marketcap indicator (above).  This should give some indication how widely held the stock is.  But one should be careful interpreting this indicator...a large cap stock like Microsoft will often have a relatively low value, since its huge marketcap obliterates the "number of institutions" figure.
12. Shorts: the short interest ratio
13. Insiders: expressed as a percentage of the company that is held by insiders

Occasionally, you might see the following indicators in our tables.  They relate to the time of the week, month, the month itself, and the quarter, just after our indicators were last current.  

1. day of week:  the day of week after our indicators were current.  Thus if the stock last traded on Friday, the number you'll see is "1", since Monday is the next trading session.
2. day of month:  as above, except here we output a number from 1 through 31, indicating the day of the month.
3. month:  as above, except here the number refers to the month in which our indicators were last current (1-12).
4. month of quarter:  as above, except here we're asking in which month of the quarter (1-3) our indicators were last current. 

Note that there are a limited number of values that can be output with the above 4 indicators.  In fact, the "month of quarter" indicator only has three values.  If we divide our underlying data into 10 "slices" (deciles) per column, how do we decide which when a value of "1" belongs in the first, second, third, or fourth decile?  The answer is that it's done randomly.

home page

    Copyright © 2007 Stockwarrants.com. All rights reserved.