Nice writeup. Pretty basic stuff, but a lot of effort. (I've worked with a company off and on that does a similar form of machine learning.)
Good results for a prototype, though I'm not sure it's ready for prime time.
M* acknowledges that the model has a hard time distinguishing between gold, silver, and bronze. Grouping these all together into one bucket (called "Recommended"), the model still doesn't do a great job. See Exhibit 4 in the paper.
Of funds that analysts actually rated Recommended (gold, silver or bronze), the model only thought 78% of them should be recommended. The model said that 21% of these Recommended funds should be rated Neutral, and 1% Negative.
Of the funds that analysts rated Neutral, the model said 59% should be Neutral, it Recommended 32% of them, and was Negative on 9%.
Of the funds that analysts rated Negative, the model rated only 55% Negative, it was Neutral on 41% and even Recommended 4%.
Given the questionable value of analyst ratings in the first place, an automated rating that is itself wrong 22% to 45% of the time seems to render this first stab as useless.
I've no doubt the accuracy can be improved. Will take a lot of time and effort though.

Interesting move by PIMCO, eh? Contrarian indicator?
Pimco Bucks Trend by Raising Fees on Ivascyn's Income Fund
Pimco Income Fund, the fastest-growing actively managed U.S. mutual fund, is raising fees at a time most money managers are racing to slash costs paid by investors.
The $92 billion fund will increase its fees 5 cents per $100 on Oct. 2 for most share classes, according to a filing Thursday, bringing expenses for the institutional-class shares to 50 cents per $100. The change in supervisory and administrative fees is attributed in part to the cost of managing dividends for the growing fund, which has about 5,500 securities in numerous countries and currencies, said Pacific Investment Management Co. spokesman Michael Reid.
< - >
https://www.bloomberg.com/news/articles/2017-08-31/pimco-bucks-trend-by-raising-fees-on-ivascyn-s-income-fund

FYI: U.S. stocks climbed Thursday after the Commerce Department said spending by U.S. consumers grew in July, along with wages and salaries. Technology companies are extending their winning streak as the Nasdaq composite nears record highs. Health care and energy companies are also rising, and so are retailers. Packaged food companies are falling after weak forecasts from Campbell Soup.
Regards,
Ted
Bloomberg:
https://www.bloomberg.com/news/articles/2017-08-30/asia-stocks-set-to-open-mixed-on-u-s-data-korea-markets-wrap
Reuters:
http://www.reuters.com/article/us-usa-stocks-idUSKCN1BB1GY?il=0
MarketWatch:
http://www.marketwatch.com/story/sp-500-set-to-edge-higher-but-facing-first-monthly-loss-since-march-2017-08-31/print
IBD:
http://www.investors.com/market-trend/stock-market-today/nasdaq-up-0-8-as-microsoft-alphabet-stocks-make-strides/
CNBC:
https://www.cnbc.com/2017/08/31/us-stocks-august-economy.html
AP:
http://hosted.ap.org/dynamic/stories/F/FINANCIAL_MARKETS_ASOL-?SITE=AP&SECTION=HOME&TEMPLATE=DEFAULT
WSJ: Markets At A Glance:
http://markets.wsj.com/us
SPDR's Sector Tracker:
http://www.sectorspdr.com/sectorspdr/tools/sector-tracker
SPDR's Bloomberg Sector Performance Pie Chart:
https://www.bloomberg.com/markets/sectors
Current Futures: Positive
http://finviz.com/futures.ashx

Thanks @Ted. Hoping not to step on any toes here. What struck me today was that commodities - notably energy (followed by gold) were on a tear. One day is probably meaningless, but just relaxing in front of the Bloomberg screen after my afternoon workout I see:
- RBOB Gas (gasoline futures) +13.5% for the day - after an already strong week
- Heating Oil + 5%
- Nat gas + 3%
- NYMEX + 2.5%
- AG - Very strong across the board
- Lumber Up
- Aluminum Up
- Copper Up
- Gold + $13 to over $1320 (about $100 above its recent low in May)
*'Footnote - Giving credit where due: @Ted's AP Link does include a pretty good summary of the energy & commodities markets, although some of the numbers differ a bit from what Bloomberg is showing.

FYI: U.S. stocks climbed Thursday after the Commerce Department said spending by U.S. consumers grew in July, along with wages and salaries. Technology companies are extending their winning streak as the Nasdaq composite nears record highs. Health care and energy companies are also rising, and so are retailers. Packaged food companies are falling after weak forecasts from Campbell Soup.
Regards,
Ted
Bloomberg:
https://www.bloomberg.com/news/articles/2017-08-30/asia-stocks-set-to-open-mixed-on-u-s-data-korea-markets-wrap
Reuters:
http://www.reuters.com/article/us-usa-stocks-idUSKCN1BB1GY?il=0
MarketWatch:
http://www.marketwatch.com/story/sp-500-set-to-edge-higher-but-facing-first-monthly-loss-since-march-2017-08-31/print
IBD:
http://www.investors.com/market-trend/stock-market-today/nasdaq-up-0-8-as-microsoft-alphabet-stocks-make-strides/
CNBC:
https://www.cnbc.com/2017/08/31/us-stocks-august-economy.html
AP:
http://hosted.ap.org/dynamic/stories/F/FINANCIAL_MARKETS_ASOL-?SITE=AP&SECTION=HOME&TEMPLATE=DEFAULT
Bloomberg Evening Briefing:
https://www.bloomberg.com//news/articles/2017-08-31/your-evening-briefing
WSJ: Markets At A Glance:
http://markets.wsj.com/us
SPDR's Sector Tracker:
http://www.sectorspdr.com/sectorspdr/tools/sector-tracker
SPDR's Bloomberg Sector Performance Pie Chart:
https://www.bloomberg.com/markets/sectors
Current Futures: Positive
http://finviz.com/futures.ashx

Do these ETFs deduct 10% and give it to the poor for tithing? I guess not:
https://thegospelcoalition.org/article/7-reasons-christians-not-required-to-tithe
or maybe:
https://washingtonpost.com/news/acts-of-faith/wp/2017/08/29/we-were-never-closed-joel-osteens-houston-megachurch-disputes-claims-it-shut-its-doors/?utm_term=.82409523c2a6
The Lord works in mysterious ways.

FYI: Making money in the markets is tricky enough on its own. Try doing it while staying faithful to your religious beliefs.
Regards,
Ted
http://www.reuters.com/article/us-religion-etf-idUSKCN1BB2JI

Ah. Thanks v much for specific example. Interesting. I bet that was not unique. That's a nontrivial bid-ask.
I just checked at ML and for AOR at the moment it's 44.24 and 44.28, 4 cents. ~1% if my arithmetic is good today. Vol 56k. I did not try and trade a small quantity, so I suppose worse is possible. I did not check at Fido.
Given the kind of entity it is, such a nice diverse mix of other etfs (AOA, AOK, and AOM are very similar, different proportions), I guess it should be buy-hold and not something to consider trading.
But your point is taken and I will remember what I have learned when I suggest it in the future. Thanks much.

FYI: Risk is a tricky concept to grasp because most people don’t really understand what their true risks are. We tend to worry about things that are readily available in our memories while spending little time worrying about problems that could be years or decades out into the future. For the majority of investors, their biggest risk is not meeting their goals, or worse, running out of money. This piece I wrote for Bloomberg looks at a study that sheds some light on another risk associated with longevity.
Regards,
Ted
http://awealthofcommonsense.com/2017/08/the-biggest-risk-for-most-retirees/

FYI: Nick Murray says “The ability to distinguish between volatility and loss is the first casualty of a bear market.”
I turned to one of my favorite passages of his masterpiece Simple Wealth, Inevitable Wealth this morning as turmoil from overseas and the commodity markets made its way through the headlines. Nick relays a great anecdote about how much money one investor personally “lost” during the last Russian Ruble crisis in the summer of 1998…
Regards,
Ted
http://thereformedbroker.com/2017/08/30/an-amazing-lesson-from-buffett-on-his-cake-day/

I am not sure why I even open up M* articles like this... Most of these funds are not really unique.
Why does M* even mention funds with a minimum of $10,000,000? Granted a few people may have access in a 401k but then they will have looked for information on their own.

@Wholly Terrier: In the future , would appreciate if would link article .
Regards,
Ted
https://corporate1.morningstar.com/ResearchLibrary/article/817415/morningstar-prospects---q2-2017/

@Wholly Terrier: In the future , would appreciate if would link article .
Regards,
Ted
https://corporate1.morningstar.com/ResearchLibrary/article/817415/morningstar-prospects---q2-2017/

Will read up on the link. My only direct experience from a year back was losing 1.5% on a market trade of 500 shares due to bid-ask spread. Limit order penny below ask did not go through for a long time and was forced to enter market order because I wanted to get out to fund another position. Never happened to me for high volume ETFs. Note that some days the volume can be dramatically lower than the average volume. Could be one-off.

Thanks @Maurice for posting this.
"Buffett’s firm pulled the trigger on a 2011 options deal that lets it buy 700 million shares — at a 70 percent discount to its share price — all for bailing out the bank when it was facing questions about its soundness."
Note that Buffett was patient and waited 6 years for his bet to pay off. And note that he invested in the bank when many considered it a "falling knife" and were fleeing. But Buffet's patience paid off and he gets the last laugh.
I saw him on Bloomberg today and, although he's 15-20 years older than me, he appears to have more energy. Yikes. All that CocaCola I guess. :)

You can see the full list here -
corporate1.morningstar.com/ResearchLibrary/article/817415/morningstar-prospects---q2-2017/
Hood River Small-Cap Growth HRSMX has been briefly mentioned on MFO a couple times. Anyone have an opinion (other than investor shares are expensive)?

If you think M-Star's current ratings are shaky, you are going to love this -
"The number of funds that receive an Analyst Rating is limited by the size of the Morningstar analyst team. To expand the number of funds we cover, we have developed a machine-learning model that uses the decision-making processes of our analysts, their past ratings decisions, and the data used to support those decisions. The machine-learning model is then applied to the "uncovered" fund universe and creates the Morningstar Quantitative Rating™ for funds (the Quantitative Rating), which is analogous to the rating a Morningstar analyst might assign to the fund if an analyst covered the fund. These quantitative ratings predictions make up what we call the Morningstar Quantitative Rating. With this new quantitative approach, we can rate nearly 6 times more funds in the U.S. market."
As far as I can tell, it is not satire.
corporate1.morningstar.com/ResearchLibrary/article/813568/morningstar-quantitative-rating-for-funds-methodology/

Pardon ignorance, but not following --- how does volume affect much of anything in this case? They are made up of a variety of other etfs ....
Ah, if you had a mil (20k sh) in AOA and wanted to sell it all at once, that would depress your price?
Is this informative?
https://sixfigureinvesting.com/2015/08/determining-liquidity-of-low-volume-etf-etn/

https://www.sec.gov/Archives/edgar/data/1500604/000119312517273155/d447144d497.htm
497 1 d447144d497.htm 497
Janus Detroit Street Trust
The Health and Fitness ETF
Supplement dated August 30, 2017
to Currently Effective Prospectus and
Statement of Additional Information (“SAI”)
The Board of Trustees of Janus Detroit Street Trust (the “Trust”) approved a plan to liquidate and terminate The Health and Fitness ETF (the “Fund”), effective on or about October 2, 2017 (the “Liquidation Date”). After the close of business on or about September 26, 2017, the Fund will no longer accept creation orders. Trading in the Fund will be halted prior to market open on or about September 27, 2017. Proceeds of the liquidation are currently scheduled to be sent to shareholders on or about October 3, 2017. Termination of the Fund is expected to occur as soon as practicable following the liquidation.
Prior to and through the close of trading on The NASDAQ Stock Market LLC (“NASDAQ”) on September 26, 2017, the Fund will undertake the process of closing down and liquidating its portfolio. This process may result in the Fund holding cash and securities that may not be consistent with its investment objective and strategies. During this period, the Fund is likely to incur higher tracking error than is typical for the Fund. Furthermore, during the time between market open on September 27, 2017 and the Liquidation Date, because shares will not be traded on NASDAQ, there may not be a trading market for the Fund’s shares.
Shareholders may sell shares of the Fund on NASDAQ until the market close on September 26, 2017 and may incur typical transaction fees from their broker-dealer. Shares held as of the close of business on the Liquidation Date will be automatically liquidated for cash at the current net asset value. Proceeds of the liquidation will be paid through the broker-dealer with whom you hold shares of the Fund. Shareholders will generally recognize a capital gain or loss on the liquidation proceeds. The Fund may or may not, depending upon the Fund’s circumstances, pay one or more dividends or other distributions prior to or along with the liquidation payments. Please consult your personal tax advisor about the potential tax consequences.
After the Liquidation Date, all references to the Fund will be deemed to have been removed from the SAI.

I was talking about the ETFs of ETFs David mentioned. Average daily trading volumes from M*
AOR: 131,000
AOK: 72,000
AOM: 125,000
AOA: 71,000
That's not a lot of volume. Of course, I am aware of the highly traded iShare and other ETFs, and have used them to get sector or market exposure.

I have minor issues with his methodology and definitions. Nevertheless, his conclusions are reasonably valid. Though rather than saying that there's an inverse relationship (negative correlation) between analyst rating and actual future performance, I'd be more inclined as to describe the relationship as none or random.
M* says that "The Analyst Rating is based on the analyst's conviction in the fund's ability to outperform its peer group and/or relevant benchmark on a risk-adjusted basis over the long term."
The author looked at raw performance, not risk-adjusted performance. So that's one issue. Another is related to the definition of analyst ratings. Obviously M* isn't really looking at a fund's ability to outperform its relevant benchmark. Else how could it give VFINX a gold rating, when with near certainty it will underpeform its relevant benchmark over any time period? M* is really rating funds against their peers.
With that in mind, a better (or at least different) analysis would be to look at the percentage of gold, silver, bronze, neutral, and negative funds that outperformed their peers over his selected five year period (1/1/2012 - 12/31/2016). Using this metric, the results are virtually independent of rating.
6/27 (22%) of large cap gold funds failed to beat their category peer average. (Two additional funds were merged out of existence: Vanguard Tax-Managed G&I, and Morgan Stanley Focus Growth).
3/15 (20%) of large cap silver funds failed to beat their peer average.
3/11 (27%) of large cap bronze funds failed to beat their peer average.
No neutral funds (out of 11 survivors) failed to beat their peers, though Columbia Value and Restructuring (UMBIX) and Putnam Voyager (PVOYX) were merged away.
The author reviews only one of the two negatively rated funds. I assume that's APGAX. Aside from a grouping of just one fund not being meaningful, this fund changed management almost immediately (Feb 2012). Curiously, M* still refuses to rate this now five star fund above neutral, because it says that five years history isn't long enough. The other negatively rated fund, the one I think the author disregarded, is LMGTX. This one also changed in 2012, even more significantly. It changed from being classified domestic to being classified foreign.
Another problem with the analysis is that because of the funds M* selected, there is a tendency to double count. It's as if M* gave medals in 1998, and awarded gold to half a dozen Janus funds, all virtual copies of each other.
That's what happened with at least a couple of families: Yacktman (where gold Yacktman and silver Focused both underperformed), and Weitz (silver Value and gold Partners Value both underperformed).
Overall, I think the best (worst?) you can say is that the analyst ratings are a better indication of which funds are popular (M* doesn't pay much attention to lesser-followed funds) than they are any indication of how funds will perform.