StockFetcher Forums · General Discussion · Would this be a good enough baseline<< >>Post Follow-up
styliten
277 posts
msg #157142
Ignore styliten
7/3/2021 3:50:16 PM

If not, what do you guys use as a preferred baseline of any stocks?

Fetcher[dow 30

set{c16, cema(Center Linear Regression Line(16, 1.0) add to Center Linear Regression Line(20, 1.0), 1) / 2}
set{c17, cema(Center Linear Regression Line(17, 1.0) add to Center Linear Regression Line(19, 1.0), 1) / 2}
set{c18, Center Linear Regression Line(18, 1.0)}

and draw price line at c16
and draw price line at c17
and draw price line at c18
]



Cheese
1,374 posts
msg #157145
Ignore Cheese
7/3/2021 9:41:22 PM

https://www.stockfetcher.com/forums/General-Discussion/Would-this-be-a-good-enough-baseline/157142
styliten 7/3/2021 3:50:16 PM
============================================================



sty,

You may find the following thread on IBD Flat Base line of added value:
https://www.stockfetcher.com/forums/General-Discussion/Flat-Base/56824/-1/56824

Other pros use ATR.
There are good ATR posts by four, davesaint86, and others

Yet other pros use the Kijun Base Line.

styliten
277 posts
msg #157162
Ignore styliten
modified
7/6/2021 1:26:48 PM

@ Cheese,

Got some questions for you:

Have you ever figured out/able to reconcile between:

a. https://www.stockfetcher.com/forums/Indicators/Linear-Regression-Channels/43 and
b. https://www.stockfetcher.com/forums/Indicators/Standard-Error-Channels/96317

Specifically, using 16 or 25 days (the square roots of 16 and 25 are 4 and 5 respectively):

1. What's the difference between:

Draw Top Linear Regression Line(16,1.0,0) and
Draw Top Linear Regression Line(16,1.0,1)

2. What's the conversion between Linear Regression Channel and Standard Error Channel?

https://www.investopedia.com/ask/answers/042415/what-difference-between-standard-error-means-and-standard-deviation.asp

Since

SEM1

Shouldn't

Draw Top Linear Regression Line(16,1.0,1) = Draw Upper Standard Error Channel (16, 4) ?

If not, what's the correct conversion value between the two using 16 or 25 days as an example?

Fetcher[Dow 30
Chart-time is 45 days

Draw Top Linear Regression Line(16,1.0,0)
Draw Upper Standard Error Channel(16, 4)
]



Cheese
1,374 posts
msg #157164
Ignore Cheese
7/6/2021 3:03:18 PM

sty,

Your questions are beyond my pay grade.

Perhaps, technical minds like nibor100, fotch, xarlor, mahkoh, and many more members
can answer questions.

I simply get down to my level and use the following link, which is good enough for me:
https://s4be.cochrane.org/blog/2018/09/26/a-beginners-guide-to-standard-deviation-and-standard-error/

styliten
277 posts
msg #157168
Ignore styliten
modified
7/6/2021 10:24:59 PM

Leaning towards the opinion that the built-in stderr calculations may be wrong:

Fetcher[
dow 30

set{t16, Top Linear Regression Line(16,1.0, 1)}
set{b16, Bottom Linear Regression Line(16,1.0, 1)}
set{u16, Upper Standard Error Band(16,1.0)}
set{l16, Lower Standard Error Band(16,1.0)}

set{t25, Top Linear Regression Line(25,1.0, 1)}
set{b25, Bottom Linear Regression Line(25,1.0, 1)}
set{u25, Upper Standard Error Band(25,1.0)}
set{l25, Lower Standard Error Band(25,1.0)}

set{t36, Top Linear Regression Line(36,1.0, 1)}
set{b36, Bottom Linear Regression Line(36,1.0, 1)}
set{u36, Upper Standard Error Band(36,1.0)}
set{l36, Lower Standard Error Band(36,1.0)}

and add column t16
and add column b16
and add column u16
and add column l16
add column cstddev(close, 16) {sd16b}
add column StdErr(close, 16) {SE16}

and add column t25
and add column b25
and add column u25
and add column l25
add column cstddev(close, 25) {sd25b}
add column StdErr(close, 25) {SE25}

and add column t36
and add column b36
and add column u36
and add column l36
add column cstddev(close, 36) {sd36b}
add column StdErr(close, 36) {SE36}
]



View the results in TABLE.

The good thing: the difference between Top & Bottom Linear Regression Line does equal to standard deviation, and the difference between Upper and lower Standard Error Band does equal to the stderr value.

The bad thing: the stderr is sometimes greater than the corresponding stddev for the same period, which is statistically impossible!

Std-Error2


Wish there is more technical know-how here!

Cheese
1,374 posts
msg #157172
Ignore Cheese
7/6/2021 10:50:52 PM

sty,

In the earlier years, members like TRO, yepher, alf44, Kevin_in_GA, miketranz, niko, etc
would be able to have a meaningful discussion.

TRO, yepher, and mahkoh have recently resurfaced. Maybe they could chime in.

Currently, members like nibor100, xarlor, miketranz, etc would likely be qualified to
provide insights.

Also, please contact sF staff. Usually, people who programme the indicators have more
insightful views than users do.

Cheese
1,374 posts
msg #157173
Ignore Cheese
7/6/2021 11:02:57 PM

sty,

At one point, your posts seemed to refer to Standard Error Channel and Standard Error Band
interchangeably. But sF explanations said they are different.

Cheese
1,374 posts
msg #157174
Ignore Cheese
7/6/2021 11:26:13 PM



For all practical purposes, Linear Regression Line and Standard Error Channel
should give the average users the same trading decision support


Fetcher[
SYMLIST(AAPL)

chart-time 4 months

/* To create the linear regression channels, we use the popular method of finding the point furthest from the least squares line and create two channels of equal width from that point */



draw Top Linear Regression Line(16)
set{TLRL, Top Linear Regression Line(16) }
add column TLRL
]







Fetcher[
SYMLIST(AAPL)

chart-time 4 months

/*
Standard Error Channels are very similar to Standard Error Bands except the channel boundaries are projected back from the most recent Standard Error and Linear Regression Indicator measures.

Where the Standard Error Bands are re-evaluated at each date, the Standard Error Channels, are linear projections based on single point to help identify the current trend based on the input parameters.
*/


draw Upper Standard Error Channel(16)
add column Upper Standard Error Channel(16)
]








Fetcher[
SYMLIST(AAPL)

chart-time 4 months

/*
Standard Error Channels are very similar to Standard Error Bands except the channel boundaries are projected back from the most recent Standard Error and Linear Regression Indicator measures.

Where the Standard Error Bands are re-evaluated at each date, the Standard Error Channels, are linear projections based on single point to help identify the current trend based on the input parameters.
*/


draw Upper Standard Error Band(16)
add column Upper Standard Error Band(16)
]







styliten
277 posts
msg #157176
Ignore styliten
7/6/2021 11:27:18 PM

Yes, Error Bands and Error Channels are different, similar to LRI and Linear Regression (Channels) respectively.

But they are identical in values as of "today" (terminal value):

Fetcher[dow 30

set{u16a, Upper Standard Error channel(16,1.0)}
set{u16b, Upper Standard Error Band(16,1.0)}
set{u16a_2, Upper Standard Error channel(16,2.0)}
set{u16b_2, Upper Standard Error Band(16,2.0)}
set{u16a_3, Upper Standard Error channel(16,3.0)}
set{u16b_3, Upper Standard Error Band(16,3.0)}

set{l16a, Lower Standard Error channel(16,1.0)}
set{l16b, Lower Standard Error Band(16,1.0)}
set{l16a_2, Lower Standard Error channel(16,2.0)}
set{l16b_2, Lower Standard Error Band(16,2.0)}
set{l16a_3, Lower Standard Error channel(16,3.0)}
set{l16b_3, Lower Standard Error Band(16,3.0)}

add column u16a
add column u16b
add column u16a_2
add column u16b_2
add column u16a_3
add column u16b_3

add column l16a
add column l16b
add column l16a_2
add column l16b_2
add column l16a_3
add column l16b_3
]



See results in table view:

errvschannel

StockFetcher Forums · General Discussion · Would this be a good enough baseline<< >>Post Follow-up

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2022 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus


This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.