Acting on a suggestion from timsk, I am posting some stuff which was posted on another thread: "Technical Analysis Price, (Volume), ...". I hope it will be helpful.
One of the problems with P and V is demonstrated in the attachment. The problem is that unless you do a "continuing" refresh of your data, what you can (and frequently do) end up with is a mess. Has anyone ever noticed a "funny" candle or OHLC/HLC bar? One that for some reason seemed to change from one trace to another. The legend for the attachment is:
(SC = Sierra Charts; IB = Interactive Brokers; TWS = Traders Workstation)
1. SC chart before refresh (to do a refresh with SC you must go the "Edit" tab, scroll to the "Edit Data" line and then set the parameters for a "delete and reload".
2. SC chart after refresh. Check the volume bars and candles closely. There are some major differences between the data sets.
3. IB TWS chart "refreshed" to compare it to the refreshed SC chart of 2. One can do a refresh with IB by simply clicking the "refresh" icon.
4. A repeat of 3.
5. A Prophet chart "refreshed" to compare it to the refreshed SC chart of 2. One can do a refresh with Prophet by clicking the "GO" icon (not shown).
Charts 2, 3 and 5 are quite similar although not identical and are completely different from Chart 1. If your time bar for trading is the 1 minute bar and you don't refresh roughly every 1 minute, you're going to end up with spurious data.
I have spoken with Prophet, IB and SC about this problem and to date the basic answer is "Deal with it". Not surprising. I don't know if this artifact is universal. Try it yourselves. If your raw data vendor has some kind of "auto-refresh" in their software you may not have this problem. I also don't know to what extent tick data is affected by this phenomenon.
For me, this artifact is a total pain but at least if one knows about it one can devise a strategy to deal with it. One can make a decision to alter one's trading strategy if it appears that this problem coyld be affecting their current protocol.
Any input from anyone would be greatly appreciated. TIA.
lj
PS: The attachment has been checked with NAV and is clean.
There are some followup comments on the other thread which may also be useful.
lj
One of the problems with P and V is demonstrated in the attachment. The problem is that unless you do a "continuing" refresh of your data, what you can (and frequently do) end up with is a mess. Has anyone ever noticed a "funny" candle or OHLC/HLC bar? One that for some reason seemed to change from one trace to another. The legend for the attachment is:
(SC = Sierra Charts; IB = Interactive Brokers; TWS = Traders Workstation)
1. SC chart before refresh (to do a refresh with SC you must go the "Edit" tab, scroll to the "Edit Data" line and then set the parameters for a "delete and reload".
2. SC chart after refresh. Check the volume bars and candles closely. There are some major differences between the data sets.
3. IB TWS chart "refreshed" to compare it to the refreshed SC chart of 2. One can do a refresh with IB by simply clicking the "refresh" icon.
4. A repeat of 3.
5. A Prophet chart "refreshed" to compare it to the refreshed SC chart of 2. One can do a refresh with Prophet by clicking the "GO" icon (not shown).
Charts 2, 3 and 5 are quite similar although not identical and are completely different from Chart 1. If your time bar for trading is the 1 minute bar and you don't refresh roughly every 1 minute, you're going to end up with spurious data.
I have spoken with Prophet, IB and SC about this problem and to date the basic answer is "Deal with it". Not surprising. I don't know if this artifact is universal. Try it yourselves. If your raw data vendor has some kind of "auto-refresh" in their software you may not have this problem. I also don't know to what extent tick data is affected by this phenomenon.
For me, this artifact is a total pain but at least if one knows about it one can devise a strategy to deal with it. One can make a decision to alter one's trading strategy if it appears that this problem coyld be affecting their current protocol.
Any input from anyone would be greatly appreciated. TIA.
lj
PS: The attachment has been checked with NAV and is clean.
There are some followup comments on the other thread which may also be useful.
lj