February 15, 2013 at 9:06 pm
Hello, I am trying to calculate a ratio, i am pretty sure the problem is in the line * SUM([SUMTOTALVOLUME])) FROM #t) *
I am trying to do the ratio by TradeHour, but judging by how small the ratio numbers in my result set [RatioVolume] (dont add up to 1.0) it looks like the
SUM([SUMTOTALVOLUME]) is across all trade hours (TradeHour) rather than the trade hour I want it to be over. What changes do i need to make to fix that? TIA
SELECT TradeHour ,AVG([NumCodes]) as [AVGNUMCODES] ,[MCLower] ,[MCUpper] ,SUM([TotalVolume]) AS [SUMTOTALVOLUME] ,(SUM([TotalVolume]) / AVG([NumCodes]) ) as [AVGVOLUME]
INTO #t FROM [dbo].[EODVolumeStats] GROUP BY TradeHour, [MCLower],[MCUpper]
SELECT #t.TradeHour ,#t.[MCLower] ,#t.[MCUpper] ,CONVERT(DECIMAL(18,4),[SUMTOTALVOLUME]/(SELECT CONVERT(DECIMAL(18,4),SUM([SUMTOTALVOLUME])) FROM #t)) as [RatioVolume] FROM #t
GROUP BY #t.TradeHour, #t.SUMTOTALVOLUME, #t.MCLower, #t.MCUpper DROP TABLE #t
February 17, 2013 at 6:21 pm
In looking at this, I don't think you need to use a temp table. However in order to show you how, you'd need to provide:
- DDL for your [EODVolumeStats] table
- Some consumable sample data
- Expected results
Having said that, if the only issue you're facing is that the summary is occuring over all trade hours of the day, and you only want it over specific hours, perhaps something like this can be used to trim the time period down to an hour.
DATEADD(hour, DATEDIFF(hour, 0, TradeHour), 0)
My thought question: Have you ever been told that your query runs too fast?
My advice:
INDEXing a poor-performing query is like putting sugar on cat food. Yeah, it probably tastes better but are you sure you want to eat it?
The path of least resistance can be a slippery slope. Take care that fixing your fixes of fixes doesn't snowball and end up costing you more than fixing the root cause would have in the first place.
Need to UNPIVOT? Why not CROSS APPLY VALUES instead?[/url]
Since random numbers are too important to be left to chance, let's generate some![/url]
Learn to understand recursive CTEs by example.[/url]
[url url=http://www.sqlservercentral.com/articles/St
February 18, 2013 at 4:36 pm
Thank you for your interest, much apreciated. Yes you are right about temp tables, but i am a mere mortal compared to some here! 🙂
TradeHour represents hours of the day
MCLower & MCUpper represent a range of Market capitalisations in millions, the ranges do not overlap
Volume represents trading volume of shares
The ratios I am trying to calculate use EOD data (both historial and intraday hour snapshots) and the number of shares a company has to calculate a ratio so elevated trading volume can be determined by mulitplication
So i need a matrix of Tradehour, MC range and a ratio multiplier
In the morning i want to know if the current volume is unusual. I know by the end of the day(comparing one day to the next), but i want to work that out intraday. Thats why am gathering stats across the market to work out what is high/low trading, because activity isnt uniform throughout the day in each market segment (blue chip, microcpap etc)
Im limited to EOD data , I have already done this with Course of Sales data (with great help from this forum BTW)
I can do this easily in .net code, but not SQL
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply