Post New

Hi All,

I am running my sons' college funds using the NAS Consolodators screen. I noticed that the GTR1 screener is not updating beyond 28-Nov-2025. Does anyone have an alternate for generating the picks? If not I will take an attept at coding it in Python and scraping it togeather, but if someone already has done it--please share?

Thanks, G.

Which version? Note that this might be an overcrowded trade. Recent results are good, but lower than QQQ.

                  Screen                     CAGR    From       To
NasLiquid_consolidators_20250616_musselmant   21   19970902  20230310
         NasMo_20231005_musselmant            21   19970902  20230310
                    N1T                       10   19970902  20230310
 
NasLiquid_consolidators_20250616_musselmant   24   20230310  20251128
         NasMo_20231005_musselmant            23   20230310  20251128
                    N1T                       37   20230310  20251128

https://gtr1.net/2013/?~NasLiquid_consolidators_20...
https://gtr1.net/2013/?~NasMo_20231005_musselmant:...
https://gtr1.net/2013/?~N1T:h1::trp%281,1%29ne-999...

====== links ======
Author: musselmant
Subject: other subset of Nadaq100 strategies than mine
Date: 08/13/25 4:34 PM
"A few public, rules-based “subset of the Nasdaq-100” studies are available that (1) use a subset of the Nasdaq-100 as a changing universe over time and (2) report ≥20% CAGR over windows of at least five years:"
https://www.shrewdm.com/MB?pid=346220863&wholeThre...

==============

Nasdaq100 stocks
ratio(sma(1,21),sma(1,200)) top 5
https://www.shrewdm.com/MB?pid=963283071

==================
5 stock strategy w/1.3X better CAGR, better beta, same sharpe as SPY
sp500.a !=null
ratio(tr(1,5),tr(1,200)) bottom 10
ratio(Mcap,float.s) bottom 5

nas100.a !=null
ratio(tr(1,5),tr(1,200)) bottom 10
ratio(Mcap,float.s) bottom 5

https://discussion.fool.com/t/5-stock-strategy-w-1...

The one I have is:

https://gtr1.net/2013/?h60::nas100.a:nenull:aprc%2...

But all of the links (including yours and the one above) when run as "screener" end at 20251128, which is to say over three months ago.

--G

You can create a googlesheets.
Finviz also has the list of stocks and momentum categories that will rank them.

https://finviz.com/screener.ashx?v=141&f=idx_ndx&o... has the list with 1 yr and 1 week return.
If you click on "ownership" it gives market cap and float.

Because of the performance of just using 1 year return I would go with that; I presented the annual performance of 10, 5 and 1 stocks both the average across start weeks and the worst of each year's start weeks, with MDD, a comparison with the Nasdaq100's etc. recently.

It is simpler to use and the track record is clear.

I generated added steps but I don't know that the added CAGR is special enough to justify them.

Because of the performance of just using 1 year return I would go with that; ...

It is simpler to use and the track record is clear.



Yup.

So far at least.

I've been running a live 10 stock monthly screen since 1/21/2025.
As of today, even after the last few bad days, it has done 25.4% CAGR.

OK guys,

I am trying to match the "official" listing, and I have both Python and Google Sheets generating the picks, but am finding that the results are really sensitive to exact formulas (which bothers me). So bear with me to help me make sure I am getting the algorithm correct, and hopefully someone can correct me. According to the GTR1 link for NAS Consolodators:

https://gtr1.net/2013/?h60::nas100.a:nenull:aprc%2...

The steps are:

step0: [NASDAQ 100 Member; lag=0 days] != null

I have checked and both my python and google sheets have the correct current NAS100 stocks (there are 101 of them).

step1: [Actual closing Price; share_lag=0 days; quote_lag=0 days] > 10

Currently all NAS100 stocks have a price > 10.

step2: [[Total Return Multiplier over 5 days; lag=0 days]/[Total Return Multiplier over 200 days; lag=0 days]] Bottom 10

Total return multiplier of 5 days should be: [prev close - close 6 trading days ago]/[close 6 trading days ago] and the same for 200 days. That is, today, Feb 6, I should be using [close for Feb 5 - close for Jan-29]/[close for Jan-29]. For example, for Apple, I have [275.91 - 258.28]/258.28 = 0.064, for the 200 day, it should be the same except that it is 201 trading days ago.

step3: [[Actual closing Price; share_lag=0 days; quote_lag=0 days]/[Average dollar-volume over 63 days; lag=0 days]] Bottom 7
step4: [[Total Return Multiplier over 5 days; lag=0 days]/[Average dollar-volume over 63 days; lag=0 days]] Bottom param0; Cash When Fewer

Using close of day prices yesterday (5-Feb-2026), I get the following list:

Ticker  Current Price  Return_5_Day  Return_200_Day  Avg_Dollar_Vol_63  Ret_5d_over_Ret_200d  Price_over_AvgDolVol  Ret_5d_over_AvgDolVol
DXCM    69.89          -0.05554      0.057177        3.74E+08           -0.97137              1.87E-07              -1.48E-10
SNPS    410.44         -0.14492      0.020082        9.89E+08           -7.2164               4.15E-07              -1.47E-10
CRWD    377.16         -0.15173      0.039896        1.22E+09           -3.80298              3.09E-07              -1.24E-10
ORLY    92.86          -0.0606       0.01783         5.4E+08            -3.39863              1.72E-07              -1.12E-10
DASH    183.86         -0.1147       0.070073        1.14E+09           -1.6368               1.62E-07              -1.01E-10
QCOM    136.3          -0.10459      0.018097        1.57E+09           -5.77924              8.66E-08              -6.64E-11
ISRG    476.32         -0.06153      0.017082        1.05E+09           -3.60202              4.52E-07              -5.84E-11

Does this match everyone? I'm finding that it is highly sensitive to 200 or 201 or 199 days back.

Thanks, G.

The last time I use 3 sources for data and all 3 disagreed on the stocks to pick; I never figured out why. I just used Robbie's tester as a result as a result. When each source uses which data, gets updates, etc. is unknown. That is one reason why simply using 1 year momentum is best: less likely dispute.

How can you use Robbie’s tester (GTR1) for the current value of a screen when it only reports data to 20251128? Am I missing something?

I don't know what he is doing or not doing. When convenient to him presumably he will tell us.
You can use finviz and compare to whatever googlesheets you are using.

Is Robbie still active here or has he given up? Are people checking to see if he is OK apart from investing or maintaining his tester?

Can anyone confirm that the GTR1 definition for TR(1,5) is [closing price yesterday - closing price 5 trading days ago]/[closing price 5 days ago]?

I can't access my old link to GTR1 glossary (page won't load), so anyone with a newer link that works please respond.

Thanks, G.

https://web.archive.org/web/20180831191405/http://...

I have saved it to a PDF file. In case of link rot.

Anyway, tr(x,y) isn't in the glossary. It's in the "for dummies" doc file.

tr(L,D) total return multiplier (equal to trp/100)
trp(L,D) total return percentage L=lag, D=duration

Thanks, Ray.

There is, in fact, link rot at the previous one (its the same link I had). Unfortunately, the tr(L,D) and trp(L,D) above does not specify if it is the most recent or the lagged date that goes in the divisor. I can see doing it either way, but mathematically it seems right that it is:

tr(L,D) = [Price(L) - Price(D)]/[Price(D)]

Can anyone else find the definition?

Thanks, G.

FYI someone whom uses Claude for AI work was able to confirm a 25% CAGR for the 5 stock one step 52-wk momentum only test I gave the detail on a while ago, and 22.6% for the 10 stock, but whatever Claude data he used only has data back to Feb. 2008 but through Jan.
2026.

Robbie's tester ends last Nov. but shows 21.2% for the 10 stock, 19% for the 5 stock.

hiphop wrote: "Can anyone else find the definition?"

tr(L,D) is the same as trm(L,D), and is the total return multiplier.

trm(1,5) = gprc:1 / gprc:6


as of 20251128:

    Ticker         NVDA
    trp:1,5      -2.015054
    tr:1,5       0.9798495
    trm:1,5      0.9798495
aprc:1 / aprc:6  0.9798494
 
      aprc          177
    aprc:1          177
    aprc:2         180.26
    aprc:3         177.82
    aprc:4         182.55
    aprc:5         178.88
    aprc:6         180.64
    aprc:7         186.52

From the gtr1 Command Translation:
trp(1,5) is [Total Return % over 5 days; lag=0 days]
tr(1,5) is [Total Return Multiplier over 5 days; lag=0 days]
trm(1,5) is [Total Return Multiplier over 5 days; lag=0 days]


http://gtr1.net/2013/?~TR_functions2026::sp1500.a:...



===================
===================

From my copy of the gtr1 Glossary
Last Modified: 8:25:22 PM 11/19/2008


trm
A field function calculating retrospective total returns (as multipliers) for each investment over a specified interval of market dates. Its syntax is as follows:

trm:[<lag_days>|<param_ref_1>],[<rs_days>|<param_ref_2>]

As shown, the function trm takes two arguments, each of which can be either a number or a parameter reference. The value of lag_days must be a non-negative integer, the value of rs_days must be a positive integer, and their sum must not exceed the retrospective limit. For each investment, the function trm uses its unsigned closing g-prices to calculate its total return over the period consisting of rs_days (or the value referenced by param_ref_2 for the variant being tested) number of market days ending lag_days (or the value referenced by param_ref_1 for the variant being tested) number of market days before the current trading date; the result is assigned to the calling field.

Note that if lag_days is zero, then total returns are measured through the close of the current trading date, which corresponds to the same g-prices at which trades take place. Also observe that since total return is calculated by simply dividing unsigned closing g-prices, it can also be obtained using the field functions gprc and ratio. For example, in the command fragment

field0=trm:10,126
field1=gprc:10
field2=gprc:136
field3=ratio:field1,field2 ...,

field0 and field3 are equal. Furthermore, the field functions trm and trp can be related using the field function linear (see the latter function's glossary definition for details).

By default, a meaningful retrospective total return is always calculated for every investment on every market date, regardless of how recently the stock associated with an investment may have begun trading. This is possible because GTR1 Linearization defines each investment's daily g-prices indefinitely into the past using the history of parent companies and, when the latter does not exist, interpolation. However, to filter out investments with inadequate actual pricing history for a conventional total return computation, use the field functions dsio or dspo.


====================================
trp
A field function calculating retrospective total returns (as percentages) for each investment over a specified interval of market dates. Its syntax is as follows:

trp:[<lag_days>|<param_ref_1>],[<rs_days>|<param_ref_2>]

As shown, the function trp takes two arguments, each of which can be either a number or a parameter reference. The value of lag_days must be a non-negative integer, the value of rs_days must be a positive integer, and their sum must not exceed retrospective limit. For each investment, the function trp uses its unsigned closing g-prices to calculate its total return as a percentage over the period consisting of rs_days (or the value referenced by param_ref_2 for the variant being tested) number of market days ending lag_days (or the value referenced by param_ref_1 for the variant being tested) number of market days before the current trading date; the result is assigned to the calling field.

Note that if lag_days is zero, then total returns are measured through the close of the current trading date, which corresponds to the same g-prices at which trades take place. Also, note that the field functions trp and trm can be related using the field function linear (see the latter function's glossary definition for details).

By default, a meaningful retrospective total return is always calculated for every investment on every market date, regardless of how recently the stock associated with an investment may have begun trading. This is possible because GTR1 Linearization defines each investment's daily g-prices indefinitely into the past using the history of parent companies and, when the latter does not exist, interpolation. However, to filter out investments with inadequate actual pricing history for a conventional total return computation, use the field functions dsio or dspo.

Thank you! That was what I was looking for.

Now everything makes sense (and if anyone wants python code to do NAS_Consolodators, let me know).

--G

To prepare for the day Robbie tires of working for free, using Claude and Tiingo we can use AI to run backtests of stock price data (I don't know what else; I haven't used it yet):

Tiingo stock price data offers a free tier, with paid "Power" plans at $30/month ($300/year) for individuals and $50/month ($499/year) for commercial, internal use.

The free tier provides basic data, while the Power plan includes increased API limits and faster access to data, with dedicated, higher-priced options for data redistribution.

Key Pricing Tiers for Stock Data:

Free Plan: Suitable for beginners, featuring limited hourly/daily requests and data access, often described as 50 symbols per hour.

Power Plan ($30/month or $300/year): Includes higher API limits (e.g., 10,000 hourly requests) and faster access.

Commercial Plan ($50/month or $499/year): Designed for internal business use with higher usage capacity.

Redistribution Plan: Starts around $250-$500/month for broader data access needs.

Norgate Data

Norgate explicitly advertises survivorship-bias-free US data.
They also support “current & past” index constituent sets in integrations (their docs describe datasets that include survivorship-bias-free constituents and handle index membership in portfolio backtests).
And there’s public commentary noting they provide historical constituents for indices including Nasdaq-100.

Norgate Data

I've used Norgate for many years. It's been very reliable. The survivorship-bias-free aspect is a reason I use it for backtesting with Amibroker.

Norgate Data provides fundamental data, specifically offering over 160 current fundamental data items (such as P/E ratio, EPS, current ratio, EBITDA) for stocks

what do you pay to use Norgate? Are you running backtests using its data and AI?

what do you pay to use Norgate? Are you running backtests using its data and AI?

$360 annually.

I use their data in combination with Amibroker for backtesting and running explorations (Amibroker's term for their screener) to get list of potential trades.

I download the data in the evening and run the explorations; takes just a few minutes.

In the past I spent a lot of time working on Amibroker code for trading systems and resulting explorations. Now AI can do most of the code creation work.

Post New

US Policy FAQ

---

Contact Shrewd'm
Contact the developer of these message boards.