Skip to main content
No. of Recommendations: 21
The guide to the screen definitions is here: http://boards.fool.com/guide-to-screen-definitions-26775200....
Screen definitions can be found here: http://boards.fool.com/screen-definitions-27306312.aspx
																									
LONG_SCREENS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
3pt_Relative_Value BGS COOP BLDR STC EGO DFIN KGC CCS CECE TTEC OTEX TPH ADM PLXS AMAT RUSHA ORCL BIG DELL FONR
78RPM LAZY LL AMRK PLUG TWMC IMMU AAWW BGS SSSS CENX DRIO EGO CRSP BIG FF SGMO AUTL ZG NFE TSCO MOH GDOT AHCO IIVI PENN
Advanced BGFV LAZY SVM CNSL GFI KGC FRTA LL ELMD MJCO RADA COOP VIPS UCTT CAMT CLFD HZO SGC EGO OSPN NUAN REGI MGIC FRHC GRBK
ARS PETS APAM SIEGY SCCO UNIT RIO PCH ADM COR HCSG GEF NUS K CVS VICI CMP RYN SLF EQH QIWI JNPR GTY GSK TM HBI
BI BGFV CECE KGC SGC EGO LL BGS NOMD BMCH RCII FIX CCS BERY TNET AN HUBG SSNC HSIC HOLX RBC CHKP
Blue_Skies EDUC LAZY COHN HBB MYRG FIZZ HEAR LAKE IBP
Bob SFM UNH CI MET CBRE TKC
CANSLIM-26 ETSY VEEV KNSL PYPL AMED ADBE
Dipstick HFC TS LUKOY SRLP NOV ALTM CAPL GLP DKL
FCF-26 PYPL AMZN AAPL FB ADRNY SAP HUM DELL BABA SIEGY MSFT CHTR HD SNE WMT UNH ACN BHP PGR DTEGY BBL NEE PG AMGN AVGO
Fried_500 MCK AAL KSS HFC DXC WBA UNM PSX PVH LNC XRX HPQ HPE MET HWM BBY ALL TPR AIZ LYB SYF HBI L DVN UHS
Gentle_Screamers BGFV PBI EXPI RST MJCO DSKE SSTK LL COOP LAD FRTA HZO KGC KSU AHCO AAWW UCTT LAZY CENT BGS QRVO PAAS SSSS ATVI
GS_Mungo ATVI SSSS PAAS QRVO BGS CENT LAZY UCTT AAWW AHCO KSU KGC HZO FRTA LAD COOP LL SSTK DSKE MJCO RST EXPI PBI BGFV
GS_Mungo_Voom ATVI SSSS PAAS CENT QRVO BGS AAWW KGC AHCO UCTT KSU LAZY FRTA HZO LAD LL COOP DSKE SSTK EXPI MJCO RST BGFV PBI
GS_PCF COOP AAWW LAZY LL DSKE AHCO KGC SSSS BGS PBI BGFV FRTA MJCO UCTT QRVO KSU ATVI SSTK HZO RST EXPI LAD PAAS
GSX BGFV SGC LL CECE KGC EGO BGS FIX CCS BMCH RCII AN TNET NOMD BERY HOLX HSIC HUBG RBC CHKP ATVI SSNC
GSX2 BGFV SGC SPWH HBI TPH BGS FIX CCS BMCH CMC RCII REGI ENSG BIG PHM AN ACA FUL MTH TNET BJ NOMD SNDR LAD BERY
High_Relative_Value TNET BGS COOP BLDR BMCH STC BERY EGO DFIN KGC NOMD CCS AN ATVI RBC ECPG CECE LL HUBG OTEX CTSH POWI TPH SNDR CMC
HighOnVolume SOGO PBI SOHU RST USX BGFV MJCO LAZY NFE OSTK HZO USAK BGS COOP AHCO AAWW GFI HL LAD DSKE FSLR KGC PSNL MTH PAAS
Incoming_Cash COOP SOHU SNEX FF UMC MAXR MOH TGB RILY HLF EGO LL BIG ADRNY SMCI HZO HUYA CCS IMXI HUM SAIL SYRS GRBK SSTK SNE
Low_Mult KGC EGO AMRK BGS DFIN COOP BIG CECE BLDR FONR STC SNEX DELL TPH CCS ADM
LowPS+ KIRK BGFV DSKE USX SOHU REGI SGRY LL VSTO COOP HOME MAXR PENN EXPI W ICLK SOGO CALX OSTK HL TRUP RUN VSLR STMP CXDO
LowPSvol PBI SGC BGFV LAZY UPS SITE FIX UCTT TNET HZO CENX PII BGS COOP AHCO LRN SKM AAWW HWKN BLDR FELE LECO BMCH STC BERY
Melange GRVY ELMD NGHC BLFS NTES VIRT PFSI MYRG IRBT COOP REGI FB ADRNY SPNS ATSG MBUU FN ALGN LRCX BABA INFY DLTR IESC BLDR SMCI
Microcap_Momentum AREC EFOI NBY CPST HDSN GNLN SLRX AZRX FSI ACHV STRM ALIM BHTG AUTO CLSN INTT BLCM SBOW EYEG SRTS GIFI JAKK FTSI TRMT APEN
Money_Flow SG COHN FUTU LAZY RGR KR CBB
Net-Nets_Grahamified NVFY FRD CCUR WSTL SYBX RELL MLND MN TUSK CVU MOV EMMS
OptiMan D PPBI PFG BOKF FITB NRG RF PB EVR TFC AMTD WU EXC CVBF WMB MTB SPG NTRS CFG SSB MPC FHN ABBV KRC MET
P/S_I_Love_You SNEX TWMC SPTN KR MYRG MUSA BIG ADRNY BJ BLDR CCS LAD HZO SMCI SOHU REGI LL FRTA MOH FIX PLXS SGC FJTSY COOP NGHC
PIH_Naked KODK OMI RST OSTK SOHU ADT VSTO PENN RUN STMP MAXR VSLR APPS LPSN NOVA AMD COOP JMIA HYMC EBS LL CALX LRN SAM HZNP
POG ETSY CWH BJ TSLA ENPH ADT EGO SFM PAAS AU NEM TSCO EBAY AMZN CIEN PENN TSM AEM CC COG MTSI QRTEA FOCS BSIG ERIC
POI COHN EQT LAZY LL UNFI AAWW USAK CWH OMI SPTN BIG HZO CNX UNIT BXC CC QRTEA ARCB MYRG COOP OTEL REGI CLW BERY DFIN
Quality_Earnings LAZY LL EDUC SGC LAD SPTN SFM HZO MTH NTGR ARCB SCPL MYRG COOP OTEL CLW BERY KR ADRNY NOMD AN ECPG ETFC HBB RCII
Rabbitt APPS DLHC SGC COOP HZO KGC JELD GFF PGTI ATEN AMKR COWN ECOM EGO UCTT MITK LOCO TPH BLDR SCPL AQUA CSV WOW PAAS FIX
RS-100 PBI BGFV DSKE CLS LAZY CENX CECE SGC DFIN RRC SGU PACK FRTA HBI IMXI TPH CC PGTI LL SPWH SGRY BLDR HZO CMC BMCH
S&P_Peg COG UNM
Shrinkage BSIG CTXS LOW EBAY NTAP CQP HRB STX LBTYA MTOR LGND AAPL QCOM ORCL BKNG ALSN LPLA BIIB MUSA FTNT MOH SEAS DVA DEO PAM
Silver_Parachute NHTC BWMX AM NNA DHT FRO DKL SRLP EARN YZCAY TGS MSB NEWT TPVG MBT SUN WF ORC SNP AB IEP DX MPLX CAPL HMLP
Small_Value ODP INSW TTMI CPSI COLL PLAB MX RGR PRPL HWKN LOCO RST DRD CNXN CSTL
Steady_Growth NMIH
Tech_Growers FVRR APPS FSLY ZM SE PDD DOCU FTCH NET SHOP ZS DAO TWLO SQ MELI COUP DOYU BILL TTD BILI CRWD TWOU DDOG OKTA SPT
Turnarounds COHN TUP BXC LEU DLX DCP MOS LPI AAN SPB CC HOV TDC OI SHYF OC NX RCM AIMC KELYA NLS PRSC MATW VSAT AEGN
Up_5% AAWW DOYU CMBM TSCO LMNX ARCB MUSA PFSI COOP SAH LOCO BCC DORM PATK PII WHR DHI RUSHA STC LBTYA BLDR ENSG SNDR ON TPH
Up5X3 NOA RGP EBMT MBIN STXB NCBS CMBM BFIN PACK IIIN FOR HONE PFNX WTBA MCBS
Value_at_the_Top REGI COOP SENEA FF BIG BIO RCII SGU NGHC STC DELL LEN ATSG LXP MUSA MTH BERY PGR HZO WHR DHI DAR SGC KMPR KR
VG-Horse EDUC DLHC SGC LEU LAZY HBB MHH FIZZ SENEA HEAR WMK LAKE APT HOV USNA ATNI
VG-Zebra EDUC LAZY HEAR HBB HOV LAKE MHH
WK_Voom LAZY PBI EDUC LL EGO HEAR SWBI SGMO VOXX ZNGA LCUT SCPL DFIN CYH AAWW LAKE CC RMNI TROX CDNA NTGR CTB W BIG BCC
Z26saTA GRBK MYRG STRL CCS IRWD
Zweig-26 MYRG CCS
SHORT_SCREENS
Current_Ratio_s VNOM FBRX PCG ALTM IRCP BLCM SHO PSXP STNG AINC NMM APOG DMRC CPLP TZOO WVE GOL AZUL ZGNX SALT GTHX RDUS VYGR KRP SAGE
Dilution_s NRBO PHGE ENLV KLR YTEN TARA BRMK CG NFH XFOR SPCE AFYA PACQ QLGN VVNT WEYL AHCO AKER IMVT LMNL APDN LRDG
Hiar_s CSTL HFFG NNN SMPL QDEL MSON GNMK KWR BYND TMUS INMD ALNY CYRX IR SILK FOLD OPTN LULU BLFS BMY TMHC ACAD CHRS CODX ABBV
Negative_FCF_s BA AGO CIT BMA X SAVE FUN HA PCG VNO AIR FLR CIM PEB SFE SU GEL ESRT WSR CXP AAT BUD RPAI ANF RPT
The_Mirror_s SALT TNP STNG NTUS PVBK BCML SHI FUNC PPC RGCO VTKLY
Print the post Back To Top
No. of Recommendations: 13
Larry,

What do you think about adding the ROE_Cash screen to your weekly SI rankings? We've had plenty of discussion about it but no final verdict. I chose my own SI based version which includes only US based non-financial stocks. I'm very gradually rebuilding my MI portfolio with this screen, having sold everything on 3/11/20. I no longer trust the old VL screens I was using.

I'd like it if you used my version of the screen. It would give me a place to verify my own weekly ranking. But you could of course choose to post a different version. The url for my version is below.

I guess you'd need the Radiscript translation to post the screen. Can anyone here easily do that translation?

Elan

http://gtr1.net/2013/?~ROE_Cash_noFinance_with_crap_filters:...
Print the post Back To Top
No. of Recommendations: 3
Larry,

What do you think about adding the ROE_Cash screen to your weekly SI rankings?


I second this. But also, might we also add Eric's version that includes a relative strength measure?

http://gtr1.net/2013/?~ROE_Cash_SImc1700rrs15:h42i126f0.4::M...

(see post 278123)

Cathy
Print the post Back To Top
No. of Recommendations: 7
Larry,

What do you think about adding the ROE_Cash screen to your weekly SI rankings?

I second this. But also, might we also add Eric's version that includes a relative strength measure?

(see post 278123)

Cathy


Elan's version is more consistent with the idea posed in Jim's original post. Also, Elan's version includes appropriate filters added since my post 278123. Here is what I believe to be Elan's version from post 278307, which I'd suggest assume the name ROE_Cash.
http://gtr1.net/2013/?~ROE_Cash_Elan_278307:h42i126f0.4::sty...


If Larry were to also include an ROE_Cash_rss screen, I would suggest it simply add the step rrs(1,252) top 15 to Elan's version. You can see by running the following URL that CAGR for this ROE_Cash_rrs screen tends to decline steadily with position size from 2 to 15. Also, retaining the hold while passing step 12 seems to help CAGR for the most part (not for position sizes 1 and slight penalties for position sizes 10 and 11).
http://gtr1.net/2013/?~ROE_Cash_rrs:h42i126f0.4::styp.a:et10...


Eric
Print the post Back To Top
No. of Recommendations: 13
elann:
I guess you'd need the Radiscript translation to post the screen. Can anyone here easily do that translation?

If the purpose is to verify rankings obtained from the GTR1 backtester, that's going to be very difficult or impossible for the following reasons:

1. SI Pro has not provided Cash Flow Per Share since 2010. It has fields that are currently called "Cash Flow Per Share", but these fields are more accurately described as "Net Change in Cash (and equivalents)", as reflected in the corresponding GTR1 field files that begin in 2010. These fields have been shown to be garbage. In order to get what is traditionally known as "Cash Flow Per Share" in SI Pro, someone will have to replicate the process I use to "recover" it from other SI Pro fields. See http://boards.fool.com/Message.asp?mid=30926707

2. The screen in question uses "derived" field files in place of SI Pro's own trailing-twelve-month (TTM) calculations due to the latter being error-prone. The method I use to calculate these TTM field files would need to be replicated. See http://boards.fool.com/Message.asp?mid=31334133

3. SI Pro has a bug in its market capitalization field that AAII is aware of, but apparently has no plans of fixing. Correcting this would require following the procedure I use, which would be extremely difficult outside of the GTR1 backtester. See http://boards.fool.com/Message.asp?mid=34140928

4. The main crap filter, step (3), which eliminates companies that have undergone certain corporate reorganizations within the 252 market days leading up to the lastest 10-Q period ending date, would be impossible to replicate outside the GTR1 backtester.

5. Any method of eliminating financial sector stocks will never exactly match the fincd12.a != 11--there is too much variation in SIC codes among different data sources.

6. Matching the styp.a = 10!11 step using SI Pro data would be impossible. For one thing, SI Pro gives country of headquarters, which is not the same as country of incorporation, as used in defining the values in styp.a. Other than the exclusion of foreign-incorporated companies, it may be possible to identify most of the other excluded security types with SI Pro data, but an exact match will be impossible.

Robbie Geary
Print the post Back To Top
No. of Recommendations: 6
Robbie is correct! For example taking just the first screen 3ptRelVal
from this weeks report. Screening on GTR1 and SIPro using radiscript
give the following results:

GTR1: BGFV FIX HZO CSIQ BGS STM COOP BLDR BMCH STC
SIPro: BGS COOP BLDR STC EGO DFIN KGC CCS CECE TTEC

Only 4 of the 10 BGS,COOP,STC and BLDR are common!

Backtesting 3ptRelVal for the last 10 years firs using the GTR1 Version
with Robbie’s extended fields and other improvements and then using my
downloaded historical SIP data and the standard radiscript definition as
posted every week.

       GTR1    SIPro
CAGR 11.6 8.6
GSD 25.6 19.0
MDD 49.3 25.3
AT 11.5 10.1


As you can see the results you may expect from running backtests on GTR1
no longer correspond to the results anyone should expect to get using
standard radiscript based screens posted every week.

The only way you can reliably expect to get what you have backtested is to subscribe to SIP, download the EOM data, retrieve the hd5 hash code and depend on GTR1 as your screener. Or subscribe to Portfolio 123 and depend on their backtester and screener.

RAM
Print the post Back To Top
No. of Recommendations: 10
As you can see the results you may expect from running backtests on GTR1
no longer correspond to the results anyone should expect to get using
standard radiscript based screens posted every week.


This is a revelation to me. Maybe it's been clear to others all this time. When we rank the so called SI-Pro screens in GTR1, we aren't really running SI-Pro screens.

There's an odd dissonance here. I get a subscription to SI-Pro to get a hash code to run the GTR1 screener. AAII gets all the money, Robbie does all the work.

Elan
Print the post Back To Top
No. of Recommendations: 15
When we rank the so called SI-Pro screens in GTR1, we aren't really running SI-Pro screens.

I wouldn’t express it quite so strongly. GTR1 is using both the current and historical SI-Pro data which they published from 1997. Unfortunately, the AAII people don’t seem to be all that concerned with the consistency of their fundamental data fields. Every couple of years they arbitrarily drop a few fields, create a few new ones and then change the way they calculate a couple. They even once changed all the company id numbers that over time distinguished between using identical tickers mnemonics. Only members of this board who are actually measuring how well this data has corresponded to profits seem to notice these changes.

Robbie has been doing an unbelievably difficult job of working around these changes, recalculating the fields where necessary and linking other data sources to the data. The most amazing part of this is that as I understand it he was a mathematician when he started with very little programming or finance skills.
. . . And yes Robbie does all the work! I don’t even believe we are paying for his computer or internet access.

RAM
Print the post Back To Top
No. of Recommendations: 10
"I don’t even believe we are paying for his computer or internet access."

Some of us do feel obligated to send him donations periodically to help pay for those expenses. I try to send him a "surprise" every Christmas.
Print the post Back To Top
No. of Recommendations: 7
I compared GTR1 results to the weekly ledgers about 4 years ago. The results were close except for a few screens. Screens like {3pt_Relative_Value} that use cash flow are very different on GTR1.

Best fit line to the data (with a R2 of 0.76) was:
CAGR_G = 85 * CAGR_L + 1.2

CAGR_L is the Weekly Ledgers from Friday 26-May-2006 to Monday 30-May-2016.
CAGR_G is GTR1 results from 20060526 to 20160527 (markets were closed on 20160530).

https://boards.fool.com/comparing-gtr1-results-and-weekly-tr...

"Overall, there is good agreement between the Rankings and GTR1, with 56% Jaccard index on average. The 12 screens affected by the AAII definition changes had an average 14% Jaccard index. The other 42 screens had an average 68% Jaccard index."
https://boards.fool.com/comparing-gtr1-picks-and-si-rankings...
Print the post Back To Top
No. of Recommendations: 7
Using the picks from the listing one month ago- post 278394.

Looking at the first five positions (some had fewer, though),
here are the average returns (sorted) with
the highs and lows of the five.

At the bottom of post, top 5 winners and bottom 5 losers stocks.

Many screens had some winners and losers- guess you have
to watch that egg basket...

Average	Max	Min	
36%	104%	1%	Turnarounds
35%	97%	7%	Rabbitt
31%	53%	3%	Up_5%
31%	97%	4%	POG
27%	112%	-18%	LowPS+
24%	36%	13%	Up5X3
23%	51%	3%	Current_Ratio_s
20%	26%	17%	S&P_Peg
19%	39%	0%	GS_Mungo_Voom
19%	38%	-4%	Blue_Skies
19%	36%	-12%	VG-Zebra
17%	36%	-7%	78RPM
16%	31%	1%	POI
16%	31%	0%	Value_at_the_Top
16%	26%	6%	Fried_500
15%	22%	9%	Z26saTA
15%	39%	3%	Small_Value
15%	45%	-4%	Dilution_s
15%	31%	1%	Low_Mult
15%	30%	-3%	Melange
14%	26%	4%	Bob
14%	36%	-2%	VG-Horse
12%	39%	-5%	Money_Flow
12%	28%	1%	LowPSvol
12%	39%	1%	Gentle_Screamers
12%	15%	7%	OptiMan
12%	41%	-5%	Dipstick
12%	39%	-5%	GS_Mungo
10%	28%	-21%	WK_Voom
10%	49%	-9%	PIH_Naked
10%	19%	-5%	GSX2
9%	26%	-8%	Steady_Growth
9%	18%	-1%	ARS
8%	39%	-5%	GS_PCF
8%	40%	-35%	Advanced
7.2%			SPY  ***********
6%	12%	3%	P/S_I_Love_You
6%	31%	-29%	Incoming_Cash
6%	14%	1%	FCF-26
6%	32%	-9%	Tech_Growers
5.2%			QQQ ************
5%	38%	-21%	RS-100
5%	31%	-5%	3pt_Relative_Value
5%	9%	-2%	CANSLIM-26
4%	10%	-2%	Quality_Earnings
4%	22%	-19%	Net-Nets_Grahamified
2%	14%	-16%	Silver_Parachute
1%	12%	-11%	The_Mirror_s
1%	9%	-9%	Negative_FCF_s
1%	12%	-13%	Hiar_s
0%	9%	-5%	GSX
-1%	5%	-5%	High_Relative_Value
-1%	15%	-8%	Shrinkage
-2%	31%	-27%	Microcap_Momentum
-3%	0%	-5%	BI
-9%	2%	-29%	HighOnVolume


Winners	
OSTK	112%
EXPI	104%
APPS	97%
CRNC	53%
GLNG	51%
	
Losers	
VHC	-21%
JOB	-26%
NETE	-27%
OSMT	-29%
SMICY	-35%
Print the post Back To Top
No. of Recommendations: 7
I did a sort on the ratio of Average Return to the difference between the Max and Min. I like consistency, so I'm drawn to the screens at the top of this list.

heink


Prev Avg Max Min Screen Avg/(Max-Min)
8 20 26 17 S&P_Peg 2.22222222
26 12 15 7 OptiMan 1.50000000
16 15 22 9 Z26saTA 1.15384615
6 24 36 13 Up5X3 1.04347826
15 16 26 6 Fried_500 0.80000000
37 6 12 3 P/S_I_Love_You 0.66666667
21 14 26 4 Bob 0.63636364
3 31 53 3 Up_5% 0.62000000
13 16 31 1 POI 0.53333333
14 16 31 0 Value_at_the_Top 0.51612903
19 15 31 1 Low_Mult 0.50000000
9 19 39 0 GS_Mungo_Voom 0.48717949
7 23 51 3 Current_Ratio_s 0.47916667
33 9 18 -1 ARS 0.47368421
39 6 14 1 FCF-26 0.46153846
20 15 30 -3 Melange 0.45454545
44 5 9 -2 CANSLIM-26 0.45454545
10 19 38 -4 Blue_Skies 0.45238095
24 12 28 1 LowPSvol 0.44444444
17 15 39 3 Small_Value 0.41666667
31 10 19 -5 GSX2 0.41666667
11 19 36 -12 VG-Zebra 0.39583333
12 17 36 -7 78RPM 0.39534884
2 35 97 7 Rabbitt 0.38888889
22 14 36 -2 VG-Horse 0.36842105
1 36 104 1 Turnarounds 0.34951456
4 31 97 4 POG 0.33333333
45 4 10 -2 Quality_Earnings 0.33333333
25 12 39 1 Gentle_Screamers 0.31578947
18 15 45 -4 Dilution_s 0.30612245
23 12 39 -5 Money_Flow 0.27272727
28 12 39 -5 GS_Mungo 0.27272727
32 9 26 -8 Steady_Growth 0.26470588
27 12 41 -5 Dipstick 0.26086957
5 27 112 -18 LowPS+ 0.20769231
29 10 28 -21 WK_Voom 0.20408163
34 8 39 -5 GS_PCF 0.18181818
30 10 49 -9 PIH_Naked 0.17241379
40 6 32 -9 Tech_Growers 0.14634146
43 5 31 -5 3pt_Relative_Value 0.13888889
35 8 40 -35 Advanced 0.10666667
38 6 31 -29 Incoming_Cash 0.10000000
46 4 22 -19 Net-Nets_Grahamified 0.09756098
42 5 38 -21 RS-100 0.08474576
47 2 14 -16 Silver_Parachute 0.06666667
49 1 9 -9 Negative_FCF_s 0.05555556
48 1 12 -11 The_Mirror_s 0.04347826
50 1 12 -13 Hiar_s 0.04000000
51 0 9 -5 GSX 0.00000000
54 -2 31 -27 Microcap_Momentum -0.03448276
53 -1 15 -8 Shrinkage -0.04347826
52 -1 5 -5 High_Relative_Value -0.10000000
56 -9 2 -29 HighOnVolume -0.29032258
55 -3 0 -5 BI -0.60000000
Print the post Back To Top