Charting confusion

Discussion in 'Trading Software' started by ljyoung, Mar 20, 2008.

  1. Is anyone aware of charting software which doesn't show this artifact (see attachment) or has a way of correcting it?

    Attachment: Time frame = 9:30-4:15; 15 min time bar. The same artifact is seen with the 5 min chart but NOT with the 10, 30 or 60 min charts.

    The point is that the slopes of the channel lines (or the trendlines if one just uses those) and/or the points of intersection with price are contextually-incorrect because of the dead space shown. The problem can corrected by moving the y-axis one "space" to the left. With a 2 min chart the correction involves moving the y-axis one space to the right.

    I appreciate that if one just collects data from 9:30 till 4:00 the "problem" is resolved. However, this will produce a different set of channels (or trendlines) and as well the "futures" data set will be incomplete.

    For people who use channels or trendlines, the CME Globex time frame suffers from obvious problems with respect to P/V dysjunctions in the after/premarkets.

  2. This problem is, IMO, non-trivial and it demands a solution and so I will propose one. I do not know if it is "right" but I do believe that it is reasonable.

    For those traders who use channels or trendlines, it is critical that one be able to use these chart constructs beyond the intraday time frame. As best I can determine the time frame for the two CME equity index future products, YM and ES, which is least likely to be subject to random walk price patterns or one-sided manipulation of the price of the product is the one from 9:30 AM till 4:00 PM EST. Any other time frame will not be as "safe" as this one simply because of liquidity effects on the relation between Price and Volume and as well the lack of interplay between the futures and equity cash markets.

    Even if a trader does not use trendlines or channels, he/she knows that there are increased risks associated with the trading of entities which have poor liquidity either inherently or because of the time of day when the trade is occurring.

    There is an additional consideration when one trades futures and that is the equity cash market and more particularly the effects that the one has upon the other, i.e., futures prices affect cash prices and vice versa. When the cash market is not in play, and in the US stock market this will be the time between 4:00 PM EST and 9:30 AM EST (the next day), the futures market will, in an immediate sense, be less constrained. Just look at what happens in the E-minis between 4:00 PM and 4:15 PM EST each day to see what I mean.

    The practical aspects of choosing this time frame for futures, as it relates to the construction of channels and trendlines, is that software discontinuity effects are eliminated, P/V relationships will be less subject to one-sided manipulation, channels can be constructed over multiple time frames and in those cases where reliable futures data are often not available, the cash values for the indices can be used an approximate surrogate for the futures. It is worthwhile to note that all "daily" OHLC values quoted for the E-minis are, not surprisingly, for the Globex time frame. So when you construct your daily, weekly, etc charts using the typical sources, you are using the Globex time frame.

    In closing let me say that I am not advocating that one should ignore what goes on Price-wise or Volume-wise in the time outside the working timeframe outlined above. Clearly the "overnight" extrema in futures prices are worth noting as are other parts of this low or erratic volume period. Each trader must decide for themselves what's real and what's more apparent than real. In this game there will always be smoke and mirrors and bottomless pits.