|
PureBytes Links
Trading Reference Links
|
Simple Japanese housewife? I don't think so, lol!!! Let me walk you
through the first line of stats for AAPL in prose :-). Under the
"Variable" heading is r1-TH, which means the variable being analysed
on this line is all instances in this 5 year sample (from Oct 98- Oct
03) of today's computed value of r1 minus tomorrow's actual high. In
other words, we're trying to quantify how close r1 is to tomorrow's
high. Under the "Mean" heading is -0.0044, which is equal to the sum
of the r1-TH values in the sample divided by the number of values in
the sample. Under the "StDev" (standard deviation) heading is 1.1599,
which measures the dispersion of the *individual* r1-TH values *in
this sample*. Under the "SE Mean" heading is 0.0327, which is the
standard error of the mean. It is an *estimate* of the variability of
the mean of r1-TH from multiple samples. Finally, the 95% confidence
interval (-0.0687, 0.0598) is an interval constructed by a procedure
which, 95% of the time, will yield an interval that contains the true
value of mu (where mu is the *population* mean of r1-TH while -0.0044
is only the mean of the single 5 year *sample* that we drew from the
population). This is why I said that looking at the distribution of
the r1-TH values in the sample would be more useful, lol. Try running
the exploration and then dump the results into a spreadsheet and graph
them to get a more intuitive feel for it.
--- In amibroker@xxxxxxxxxxxxxxx, Yuki Taga <yukitaga@xxxx> wrote:
> Hi Mark,
>
> Sunday, October 26, 2003, 1:26:34 PM, you wrote:
>
> M> That's a great question. I see that Pal responded with his usual
home
> M> spun wisdom and think the following is closer to what you're
looking
> M> for. I wrote an exploration to compare the support & resistance
> M> levels (s1, s2, r1, r2) to the lows and highs of the days
immediately
> M> following. The code is simple and self explanatory. You can
then
> M> copy and past the results into your favorite spreadsheet or
statistics
> M> program. I think it's most useful to look at distributions of
the
> M> results but since the graphics don't post here anymore, I did
some
> M> stats to include 95% t confidence intervals on 5 years on AAPL,
INTC,
> M> and MSFT and the last year for all of the ND100 stocks. Looks
like r1
> M> and s1 are much better than r2 and s2.
>
> M> The Exploration:
>
> M> p = (H+L+C)/3;
> M> r1 = (2*p)-L;
> M> s1 = (2*p)-H;
> M> r2 = p +(r1 - s1);
> M> s2 = p -(r2 - s1);
>
> M> Filter = 1;
>
> M> AddColumn( DateTime(), "Date", formatDateTime );
>
> M> AddColumn( r1, "r1", 1.2 );
>
> M> AddColumn( r2, "r2", 1.2 );
>
> M> AddColumn( Ref(H,1), "Tom High", 1.2 );
>
> M> AddColumn( r1-Ref(H,1), "r1-TH", 1.2 );
>
> M> AddColumn( r2-Ref(H,1), "r2-TH", 1.2 );
>
> M> AddColumn( s1, "s1", 1.2 );
>
> M> AddColumn( s2, "s2", 1.2 );
>
> M> AddColumn( Ref(L,1), "Tom Low", 1.2 );
>
> M> AddColumn( s1-Ref(L,1), "s1-TL", 1.2 );
>
> M> AddColumn( s2-Ref(L,1), "s2-TL", 1.2 );
>
> M> The Stats:
>
> M> TH = Tomorrow's High and TL = Tomorrow's Low
>
> M> 1. AAPL Oct 98- Oct 03
>
> M> Variable Mean StDev SE Mean 95.0 % CI
> M> r1-TH -0.0044 1.1599 0.0327 ( -0.0687, 0.0598)
> M> r2-TH 0.6864 1.3892 0.0392 ( 0.6094, 0.7633)
> M> s1-TL -0.0048 1.1422 0.0322 ( -0.0681, 0.0584)
> M> s2-TL -1.3852 1.7022 0.0480 ( -1.4794, -1.2909)
>
> M> 2. INTC Oct 98- Oct 03
>
> M> Variable Mean StDev SE Mean 95.0 % CI
> M> r1-TH -0.0295 0.9505 0.0268 ( -0.0821, 0.0231)
> M> r2-TH 0.7002 1.1300 0.0319 ( 0.6377, 0.7627)
> M> s1-TL -0.0293 1.0275 0.0290 ( -0.0862, 0.0275)
> M> s2-TL -1.4663 1.4852 0.0419 ( -1.5485, -1.3841)
>
> M> 3. MSFT Oct 98- Oct 03
>
> M> Variable Mean StDev SE Mean 95.0 % CI
> M> r1-TH -0.0060 0.7022 0.0198 ( -0.0449, 0.0329)
> M> r2-TH 0.5625 0.8216 0.0232 ( 0.5171, 0.6080)
> M> s1-TL -0.0057 0.7449 0.0210 ( -0.0469, 0.0355)
> M> s2-TL -1.1369 1.0378 0.0293 ( -1.1943, -1.0794)
>
> M> 4. ND100 Stocks Oct 02- Oct 03
>
> M> Variable Mean StDev SE Mean 95.0 % CI
> M> r1-TH -0.02157 0.62947 0.00395 (-0.02930,-0.01383)
> M> r2-TH 0.47090 0.77880 0.00488 ( 0.46133, 0.48047)
> M> s1-TL -0.02238 0.60138 0.00377 (-0.02977,-0.01499)
> M> s2-TL -1.02103 0.95038 0.00596 (-1.03271,-1.00935)
>
> This is super, and I really appreciate it. Now, could you (or
> anyone) tell a simple Japanese housewife with little mathematical
> background exactly what you have done here -- in prose -- with this
> 95 percent business? ^_^ Seriously, and perhaps sadly, I don't
really
> understand exactly what some of these numbers mean. I can catch on
> pretty quick I think, but I need some kind of guide so I can figure
> it all out.
>
> Thanks again!
>
> Yuki
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/GHeqlB/TM
---------------------------------------------------------------------~->
Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|