Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.soft-sys.math.mathematica > #15432 > unrolled thread

FindRoot

Started bybruce.colletti@gmail.com
First post2013-09-14 09:49 +0000
Last post2013-09-15 10:59 +0000
Articles 3 — 2 participants

Back to article view | Back to comp.soft-sys.math.mathematica


Contents

  FindRoot bruce.colletti@gmail.com - 2013-09-14 09:49 +0000
    Re: FindRoot "Dr. Wolfgang Hintze" <weh@snafu.de> - 2013-09-15 10:59 +0000
    Re: FindRoot bruce.colletti@gmail.com - 2013-09-15 10:59 +0000

#15432 — FindRoot

Frombruce.colletti@gmail.com
Date2013-09-14 09:49 +0000
SubjectFindRoot
Message-ID<l11bf5$iss$1@smc.vnet.net>
Re 9.0.1 under Windows 7.

The code below is extracted from another source.

Part I below returns 289 even though n is never given a value.  Why does the module return a value and in particular, what does 289 mean?

Part II doesn't return a value (as expected).  I don't see why Part I returns a value but not Part II.

Thanks.

Bruce


In[1]:= 
(* Part I *)
f[n_,p1_,p2_]:=Sqrt[2n](ArcSin@Sqrt[p1-1/(2n)]-ArcSin@Sqrt[p2+1/(2n)]);

myModule1[\[Alpha]_,\[Beta]_]:=Module[{A=ConstantArray["",{9,9}],n,powerEST,p1=0.2,p2=0.1},
If[
p1>p2,
powerEST=SurvivalFunction[NormalDistribution[f[n,0.2,0.1],1],Quantile[NormalDistribution[],1-\[Alpha]]];

A[[IntegerPart[10(p1-0.1)],IntegerPart[10p2]]]=Ceiling[n/.FindRoot[powerEST-(1-\[Beta]),{n,40}]]
]
];

myModule1[0.05,0.05]
Out[3]= 289
In[4]:= 
(* Part II *)
myModule2[\[Alpha]_,\[Beta]_]:=Module[{A=ConstantArray["",{9,9}],n},
SurvivalFunction[NormalDistribution[f[n,0.2,0.1],1],Quantile[NormalDistribution[],1-\[Alpha]]]
];

myModule2[0.05,0.05]
Out[5]= 1/2 Erfc[(1.64485 -Sqrt[2] Sqrt[n$272] (ArcSin[Sqrt[0.2 -1/(2 n$272)]]-ArcSin[Sqrt[0.1 +1/(2 n$272)]]))/Sqrt[2]]

[toc] | [next] | [standalone]


#15443

From"Dr. Wolfgang Hintze" <weh@snafu.de>
Date2013-09-15 10:59 +0000
Message-ID<l143u1$pr5$1@smc.vnet.net>
In reply to#15432
Am Samstag, 14. September 2013 11:49:25 UTC+2 schrieb bruce.c...@gmail.com:
> Re 9.0.1 under Windows 7.
> 
> 
> 
> The code below is extracted from another source.
> 
> 
> 
> Part I below returns 289 even though n is never given a value.  Why does the module return a value and in particular, what does 289 mean?
> 
> 
> 
> Part II doesn't return a value (as expected).  I don't see why Part I returns a value but not Part II.
> 
> 
> 
> Thanks.
> 
> 
> 
> Bruce
> 
> 
> 
> 
> 
> In[1]:= 
> 
> (* Part I *)
> 
> f[n_,p1_,p2_]:=Sqrt[2n](ArcSin@Sqrt[p1-1/(2n)]-ArcSin@Sqrt[p2+1/(2n)]);
> 
> 
> 
> myModule1[\[Alpha]_,\[Beta]_]:=Module[{A=ConstantArray["",{9,9}],n,powerEST,p1=0.2,p2=0.1},
> 
> If[
> 
> p1>p2,
> 
> powerEST=SurvivalFunction[NormalDistribution[f[n,0.2,0.1],1],Quantile[NormalDistribution[],1-\[Alpha]]];
> 
> 
> 
> A[[IntegerPart[10(p1-0.1)],IntegerPart[10p2]]]=Ceiling[n/.FindRoot[powerEST-(1-\[Beta]),{n,40}]]
> 
> ]
> 
> ];
> 
> 
> 
> myModule1[0.05,0.05]
> 
> Out[3]= 289
> 
> In[4]:= 
> 
> (* Part II *)
> 
> myModule2[\[Alpha]_,\[Beta]_]:=Module[{A=ConstantArray["",{9,9}],n},
> 
> SurvivalFunction[NormalDistribution[f[n,0.2,0.1],1],Quantile[NormalDistribution[],1-\[Alpha]]]
> 
> ];
> 
> 
> 
> myModule2[0.05,0.05]
> 
> Out[5]= 1/2 Erfc[(1.64485 -Sqrt[2] Sqrt[n$272] (ArcSin[Sqrt[0.2 -1/(2 n$272)]]-ArcSin[Sqrt[0.1 +1/(2 n$272)]]))/Sqrt[2]]

In part 1 n is in fact assigned a value, viz. the initial values 40 in FindRoot.
Having found the root close to 289 Returns this value from the module.

Best regards,
Wolfgang

[toc] | [prev] | [next] | [standalone]


#15444

Frombruce.colletti@gmail.com
Date2013-09-15 10:59 +0000
Message-ID<l143ul$pr7$1@smc.vnet.net>
In reply to#15432
Folks, please ignore my post.  I've discovered the problem:  my printer made a semicolon look like a comma -- and we know the difference this makes incoding!  I stressed over the printed code and not the displayed one.  Problem has been solved.  Sorry for the false alarm.

[toc] | [prev] | [standalone]


Back to top | Article view | comp.soft-sys.math.mathematica


csiph-web