Jumat, 02 Juni 2017

[smf_addin] Digest Number 4047

5 Messages

Digest #4047
2a
ETF Data Slippage by "Michael Thomas" thomas91112
2d
Re: ETF Data Slippage by "Randy Harmelink" rharmelink

Messages

Fri Jun 2, 2017 9:52 am (PDT) . Posted by:

rus101

Yes, ETFs do have some misfires. Frankly, I'm mostly interested in the St. Deviation and Sharpe numbers. My spreadsheets are mix of MS and Yahoo data for funds as some things are just too hard to pull from MS. Also I use http://www.etf.com http://www.etf.com data for ETFs that you helped me with before.

And yes, I did notice the wierd 1.0.2 in item numeration, the spreadsheet adjusts for that - I pulled it up in the browser and made sure the spreadsheet matches it...

Fri Jun 2, 2017 10:28 am (PDT) . Posted by:

"Michael Thomas" thomas91112

Follow about 36 ETFs on a monthly basis, notice that it appears the SMF pricing data is off by one month for many, if not all, ETFs.  For example, month-end values appear as follows:
SPY - S&P 500 ETF

                Yahoo Finance            SMF        

March 31 2017        235.74            238.08April 28 2017        238.08            241.44May 31 2017          241.44            243.36
June 2017 MTD        243.36 (6/1/17)   243.96                     243.96 (6/2/17)
Have checked with other pricing services besides YF and they show similar discrepencies vs SMF.  Note, the SMFGetYahooHistory function being used is in the same cells as the former RCHGetYahooHistory, they were not moved upward one month when the update to SMF was done recently.
Thanks for your help.

Fri Jun 2, 2017 10:47 am (PDT) . Posted by:

petekoch

I have also noticed differences in the prices generated by smfGetYahooHistory and the results returned by the Yahoo historical data page. However, I DO NOT notice a pattern of the prices being off by a month.

Another oddity is for CEFs. Although the selling price of a CEF is off, the NAV prices appears to be accurate, although I wouldn't swear to that.

BTW, the typical NAVTicker of a CEF is "XCEFX", ie, a 3-character ticker is wrapped with a beginning X and an ending X, although that is not always the case.

Fri Jun 2, 2017 10:58 am (PDT) . Posted by:

petekoch

Access to the Yahoo historical data page I mentioned is not obvious on the main Finance page. In fact, I can't find a way to get there from Finance. Here's the link I used. I wonder if it's no longer supported. The link for IXIC, e.g., is:

https://finance.yahoo.com/quote/^ixic/history?ltr=1 https://finance.yahoo.com/quote/%5Eixic/history?ltr=1 or, as copied

https://finance.yahoo.com/quote/%5Eixic/history?ltr=1 https://finance.yahoo.com/quote/%5Eixic/history?ltr=1

Fri Jun 2, 2017 3:10 pm (PDT) . Posted by:

"Randy Harmelink" rharmelink

​I'm confused -- I don't see anything off by a month. How are you getting
the month-end closing prices out of smfGetYahooHistory(), that they are off
by a month?

Here's what I get using smfPricesByDates() for your dates, and they match
your first column of numbers:

*Date* *SPY*
2017-03-31 $235.74
2017-04-28 $238.08
2017-05-31 $241.44
2017-06-01 $243.36
2017-06-02 $244.14

That pretty much matches what I see for smfGetYahooHistory("SPY",,,"m"):

Date Open High Low Close Volume
2017-06-02 $243.42 $244.35 $243.08 $244.17 88491869
2017-06-01 $241.97 $243.38 $241.64 $243.36 66864400
2017-05-01 $238.68 $242.08 $235.43 $241.44 1492547000
2017-04-01 $235.80 $239.53 $232.51 $238.08 1549613700
2017-03-01 $237.35 $239.27 $230.60 $234.71 1921474100
2017-02-01 $226.53 $236.27 $225.83 $235.43 1365136600

March is different because of an ex-dividend date during the month.

​I would trust smfPricesByDates() more than I would trust monthly data out
from smfGetYahooHistory(), ​because that function can only apply dividend
adjustments to monthly data points -- so it's adjustments will vary from
dividend adjustments made to daily data points. At this point, I wouldn't
suggest doing another other than daily time periods with
smfGetYahooHistory(), at least until Yahoo starts making dividend
adjustments.

Here is what I get for smfPricesBetween() for "SPY" during the month of
March:

*Open Date* *Open Price* *High Date* *High Price* *Low Date* *Low Price* *Close
Date* *Close Price* *Volume* *Previous Close*
2017-03-01 $237.36 2017-03-01 $239.28 2017-03-27 $231.61 2017-03-31 $235.74
1921474100 $235.45

The prices doesn't match the monthly output for March from
smfGetYahooHistory(), because of the dividend adjustment during the month.​



On Fri, Jun 2, 2017 at 10:22 AM, Michael Thomas thomas91112@
​...
wrote:

>
> Follow about 36 ETFs on a monthly basis, notice that it appears the SMF
> pricing data is off by one month for many, if not all, ETFs. For example,
> month-end values appear as follows:
>
> SPY - S&P 500 ETF
>
> Yahoo Finance SMF
>
> March 31 2017 235.74 238.08
> April 28 2017 238.08 241.44
> May 31 2017 241.44 243.36
>
> June 2017 MTD 243.36 (6/1/17) 243.96
> 243.96 (6/2/17)
>
> Have checked with other pricing services besides YF and they show similar
> discrepencies vs SMF. Note, the SMFGetYahooHistory function being used is
> in the same cells as the former RCHGetYahooHistory, they were not moved
> upward one month when the update to SMF was done recently.
>
> Thanks for your help.
>
For the Add-in, Documentation, Templates, Tips and FAQs, visit http://ogres-crypt.com/SMF

Tidak ada komentar:

Posting Komentar