Seeking Alpha's Quant Performance

Does anyone have experience with Seeking Alpha’s Quant Performance, or with other similar sites.

https://seekingalpha.com/backtest/quant

  • does Alpha’s Quant work?
  • is turnover high?
  • it is practically possible to use Alpha’s Quant in deciding whether to buy a stock
  • does Alpha’s Quant only help to give good recommendations on purchases, but not for selling?

What about;
https://www.valuengine.com/ve/VERatingPerformance
https://www.tipranks.com/stocks/msft/forecast
https://www.stockrover.com/stock-ratings/

Hi Marchus,

I am reading your posts with interest and wondering what, if anything, I might want to include in a port using InList. Again, credit to Georg who has expanded Piggyback options available to P123 members too.

It is my understanding that members outside of the United States can have some trouble using Fidelity. My apologies if this is the case for you–possibly making this an idea that might not have any potential.

But for my own trading I am only interested in ideas that have out-of-sample data and are not cherry-picked.

Fidelity’s site has A LOT OF PROBLEMS. But you can download up to 500 tickers at a time. If you sort A-Z then Z-A (and merge them) you can get 1000 tickers easily. More if you divide the downloads, using market-cap for example, and merge the spreadsheets. So it has some potential for being used in InList: if there is any good data in there.

Investars covers the analysts on Fidelity’s site. And Starmine has a way of merging the analysts opinions (a method that may or may not be useful).

Did I mention that there are a lot of problems with this site? The opinions are always delayed. There seem to be frequent errors or discrepancies in the Strarmine Ratings. But there is SOME out-of-sample data for each of the analysts ratings performance. Again, their ratings can be downloaded.

With that long background let me show you some of the out-of-sample results as you probably do not have access to Fidelity now. The first is the Starmine rating which aggregates all of the analysts’ ratings (from 8/31/2009) and runs a monthly rebalanced portfolio out-of-sample. Obviously, I would avoid large-caps if you use Stamina’s data.

The second is Investor’s data on the individual analysts. Unfortunately this is only for 2 years and they split out each sectors results. But one can get a feel for which analysts have been doing better than the others for the last 2 years. This Investars data is rebalanced monthly as is Starmine’s.

This may not be helpful but wanted to share with you while you are posting on this subject.

I may be giving away a secret–should I end up using some of this. But it is a secret that a lot of people already have access to. And, I think sites like Zacks work–in part–because they are public and many people follow the recommendations which increases the demand for the tickers in their ratings (driving up the price). A type of self-fulfilling prophesy which may be present for any useful sentiment data. So maybe the more public this is the more useful it becomes.

Jim



As a newish user to P123, and a SA subscriber, I have found it useful to try and duplicate the SA quant rating within P123 as an exercise in learning P123. All the factors used in SA are neatly laid out for you. I think its a good foundation for a ranking system for a new user.

abwillingham, I was wondering if you have managed to copy the rating system Alpha, and if the return in your case is close to what they claim? :slight_smile:

Hello, Jrinne. Could you perhaps write more about Fidelity?

  • Which rating system, in their opinion, provides the best return?

  • Can the information be applied in real-world trading?

  • Do you have access to all of these research sites through Fidelity, or do you have to have access to them all?

  • Based on the chart below, it appears that Zacks and Verus have produced the best results?

  • From what I understand of the overview, it only pertains to the life signal, not backtesting, ore?

Otherwise, I’m interested in any information you have, as well as learning more about how you apply Fidelity practically. :slight_smile:

Here is what Fidelity offers as an opinion. This is copied from a Q3 2021 report by “Integrity Research Associates” with “performance data by Investars” on Fidelity’s site:

[i]"First Place (tie) – Refinitiv/Verus had the best buy recommendation performance over all periods and the second-best buy-sell performance (except for the 1-year period where it was the top performer). It was the best sector level performer. It had the best performance in the Industrials, Consumer Discretionary, Materials and Information Technology sectors over the last 12 months.

First Place (tie) — Zacks Investment Research had the best buy-sell recommendation performance (except for the 1- year period), the second-best sector-level performance and the second-best buy performance over all periods. It was the top performer in the Energy and Healthcare sectors over the last 12 months."[/i]

I DO NOT think the above is a monthly rebalance however. Rather I think is done as per usual by Investars where stocks are added and subtracted from an “index” each day. In that regard it is similar to SA’s method–which is rebalanced daily.

Note: SA’s performance data is for a DAILY REBALANCE and I do not see anything about turnover or slippage. Also I am uncertain as to whether you can download a lot of data to be uploaded into P123 using InList or not.

You can find data for the last 2 years in each sector for each of Fidelity’s analyst for a MONTHLY REBALANCE in the image previously provided above and make you own assesment. Refinitiv/Verus and Zacks do well here also as you can see.

As I said above–WITH CAUTION–the analysts opinions can be downloaded and uploaded into P123 and used with InList. So yes.

With caution because there are always delays and errors that you will need to check and/or go to a separate site for confirmation (e.g., Zacks). Refinitv/Verus data becomes accurate on Wednesdays for screening and download. To become delayed again on the following Monday and the screen is inaccurate on Tuesday but you can correct that by look at the report manually for each ticker on Tuesday. Verus updates weekly.

To the best of my knowledge, Zacks data never becomes accurate on the Fidelity site.

Still a work in progress. I am not sure what data I will use or if it will be from Fidelity. But buy recommendations could be put into the buy rules of a port thru InList. And sell recommendations into the sell rules in a similar manner. All of these stocks could be ranked with whatever ranking system one uses now. One could keep a RankPos > X if one has one in their port already.

Thanks.

Jim

I have not finished copying it. I just casually pick away at it when I’m not distracted by shiny objects.
I am not sure how I could test it against the SA quant since I have not way to backtest the SA quant.

  • abwillingham

I tested their momentum approach: (https://seekingalpha.com/symbol/CCS/momentum/performance)

100%AlphaQ Sector
8.33%Rel%Chg(63,#sector)ƒ(x)
16.67%Rel%Chg(126,#sector)ƒ(x)
33.33%Rel%Chg(189,#sector)ƒ(x)
41.67%Rel%Chg(252,#sector)ƒ(x)

The problem is that I dont know how much weight they assign to each factor.