Tweet
I'm observing an odd problem with Intra-Day charts for BB's, Pinks, etc. So far, I haven't seen the problem on full DOW or Nasdaq, but the problem may be occuring with other than simple OTC stock charts and needs further checks for those relying on accuracy for trading.
In checking the number of intra-day periods displayed, I'm seeing more periods than should exist. The regular session trading day at the exchanges is 6.5 hours, so any period that equally divides the 390 minute trading day should have a whole number of periods (per LC1 which was alway smack on target).
Example Stock: BCON
Standard Selections....
For 30 minute, I'm seeing 14 periods, i.e. 1 extra.
For 15 minute, I'm seeing 27 periods, i.e. 1 extra.
For 10 minute, I'm seeing 40 periods, i.e. 1 extra.
For 5 minute, I'm seeing 79 periods, i.e. 1 extra.
Custom Periods.....
For 65 minute, I'm seeing 7 periods, i.e. 1 extra.
For 78 minute, I'm seeing 6 periods, i.e. 1 extra.
For 130 minute, I'm seeing 4 periods, i.e. 1 extra.
For 195 minute, I'm seeing 3 periods, i.e. 1 extra.
My best guess.....the extra period appears to be added at close, and I'm seeing some post-session trades that contributes to that observation.
These extra periods appear to be leading to incorrect moving averages on the intra-day as well. This is a show-stopper, as I rely on the accuracy of MA (long and short) for trading signals, i.e. I can no longer use the charts.
The problem is very difficult to spot because of the very odd and inconsistent increments displayed and lack of periodic vertical bars, which makes interpreting data very difficult. Using volume bars does assist in spotting the problem (which I did on BCON). "All Sessions" is off on my charts, i.e. I shouldn't be getting any pre or post market bleed-over. However, I believe I am seeing some that's integrated with these errors.
Can anyone else please confirm my observations. Thanks.
Edit1: Strike my comments about not happening on DOW or Nasdaq. It appears to be a pervasive problem. Checking a 78 period on Google (GOOG), Deere (DE), and Ford (F), they all show 6 periods in an actual 5 period day.
Edit2: I've examined the 50 Day Moving Average on Google on both a Day (D) and Half-Day (195) chart.
The 50 Day on (D) is currently ~484. The 50 Day on (195) using 100 as number of intra-day periods is ~504 (too far off to be interpolitive data error).
I then looked at a 5 minute chart for stocks with ticks in that extra period using a 1 for Moving Average, i.e. to see where the line would go. They all are indicating the data is used.
Conclusion: All Intra-Day Charts are inaccurate of their Moving Averages with the longer time period and MA's having the highest bias due to the inclusion of non-existent (extra) trading period data. As example, a 50 Day MA on Google in an intra-day chart period of 195 is currently only assessing 100 periods, i.e. the last 33.33 trading days, and 1/3 of the data is incorrect or bias to the calculation for a dual-close (or after market trade which tend to be more out-of-bounds).
In checking the number of intra-day periods displayed, I'm seeing more periods than should exist. The regular session trading day at the exchanges is 6.5 hours, so any period that equally divides the 390 minute trading day should have a whole number of periods (per LC1 which was alway smack on target).
Example Stock: BCON
Standard Selections....
For 30 minute, I'm seeing 14 periods, i.e. 1 extra.
For 15 minute, I'm seeing 27 periods, i.e. 1 extra.
For 10 minute, I'm seeing 40 periods, i.e. 1 extra.
For 5 minute, I'm seeing 79 periods, i.e. 1 extra.
Custom Periods.....
For 65 minute, I'm seeing 7 periods, i.e. 1 extra.
For 78 minute, I'm seeing 6 periods, i.e. 1 extra.
For 130 minute, I'm seeing 4 periods, i.e. 1 extra.
For 195 minute, I'm seeing 3 periods, i.e. 1 extra.
My best guess.....the extra period appears to be added at close, and I'm seeing some post-session trades that contributes to that observation.
These extra periods appear to be leading to incorrect moving averages on the intra-day as well. This is a show-stopper, as I rely on the accuracy of MA (long and short) for trading signals, i.e. I can no longer use the charts.
The problem is very difficult to spot because of the very odd and inconsistent increments displayed and lack of periodic vertical bars, which makes interpreting data very difficult. Using volume bars does assist in spotting the problem (which I did on BCON). "All Sessions" is off on my charts, i.e. I shouldn't be getting any pre or post market bleed-over. However, I believe I am seeing some that's integrated with these errors.
Can anyone else please confirm my observations. Thanks.
Edit1: Strike my comments about not happening on DOW or Nasdaq. It appears to be a pervasive problem. Checking a 78 period on Google (GOOG), Deere (DE), and Ford (F), they all show 6 periods in an actual 5 period day.
Edit2: I've examined the 50 Day Moving Average on Google on both a Day (D) and Half-Day (195) chart.
The 50 Day on (D) is currently ~484. The 50 Day on (195) using 100 as number of intra-day periods is ~504 (too far off to be interpolitive data error).
I then looked at a 5 minute chart for stocks with ticks in that extra period using a 1 for Moving Average, i.e. to see where the line would go. They all are indicating the data is used.
Conclusion: All Intra-Day Charts are inaccurate of their Moving Averages with the longer time period and MA's having the highest bias due to the inclusion of non-existent (extra) trading period data. As example, a 50 Day MA on Google in an intra-day chart period of 195 is currently only assessing 100 periods, i.e. the last 33.33 trading days, and 1/3 of the data is incorrect or bias to the calculation for a dual-close (or after market trade which tend to be more out-of-bounds).
Comment