小代码求教 - 经管之家

2012-11-23 16:37:45

data=read.csv("sysfit_11.csv")

logistInit <- function(mCall, LHS, data) {
   xy <- sortedXyData(mCall[["x"]], LHS, data)
   if(nrow(xy) <3) {
      stop("Too few distinct input values to fit a logistic")
   }
   a <- max(abs(xy[,"y"]))
   if (a != max(xy[,"y"])) a <- -a  # negative asymptote
   b <- NLSstClosestX(xy, 0.5 * a)
   value <- c(a, b)
   names(value) <- mCall[c("a", "b")]
   value
}

logist <- selfStart(~ a*x^b,initial = logistInit, parameters = c("a", "b"))
getInitial(TOTAL ~ logist(DBH,a, b), data)

#       a       b
#833.11000  20.76035

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

xingzhaoh

2012-11-23 16:47:43

epoh 发表于 2012-11-23 16:37
data=read.csv("sysfit_11.csv")

logistInit

看懂了，谢谢，如何用R做贝叶斯先验估计的95%分位点值的帖子询问可以给指点一下吗？

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

xingzhaoh

2012-11-23 16:51:17

epoh 发表于 2012-11-23 16:37
data=read.csv("sysfit_11.csv")

logistInit

看懂了，谢谢

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

2012-11-23 23:04:42

epoh 发表于 2012-11-23 16:37
data=read.csv("sysfit_11.csv")

logistInit

epoh老师，您好！
以上程序中LHS和DBH是什么意思？

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

2012-11-24 08:35:44

zhangtao 发表于 2012-11-23 23:04
epoh老师，您好！
以上程序中LHS和DBH是什么意思？

LHS : left-hand side
   simple formula log(y1) ~ x1 + x2 | I(x1^2)
   which has a single response log(y1) on the LHS
   and two parts on the RHS, separated by |.

变量DBH,TOTAL都是楼主的数据

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

点击查看更多内容…

2012-11-24 23:00:32

epoh 发表于 2012-11-24 08:35
LHS : left-hand side
simple formula log(y1) ~ x1 + x2 | I(x1^2)
which has a single ...

epoh老师，您好！
您看以下错误是什么原因？如何解决？
非常感谢！

>> randn('state',0)             % Start from a known state.
x = randn(1000,1);             % 1000 Gaussian deviates ~ N(0,1).
y = filter(1,[1 -0.6 0.08],x);  % Create a stationary AR(2)
                              % process.
[PartialACF, Lags, Bounds] = parcorr(y , [] , 2); % Compute the
                              % partial ACF with 95 percent
                              % confidence.
[Lags, PartialACF]
Bounds
parcorr(y , [] , 2)          % Use the same example, but plot
                           % the partial ACF sequence with
                           % confidence bounds.
??? Error using ==> lagmatrix
lagmatrix: wrong # of input arguments

Error in ==> parcorr at 151
X       =  lagmatrix(Series , [1:nLags]);

>>

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

2012-11-25 09:18:09

zhangtao 发表于 2012-11-24 23:00
epoh老师，您好！
您看以下错误是什么原因？如何解决？
非常感谢！

lagmatrix.m 有兩個
一個在ucsd_garch toolbox
另一個在 James P. LeSage Econometrics Toolbox, 你要用這個
http://www.spatial-econometrics.com/arfima/lagmatrix.m

%%%%%%%%%%%%%%%%%%%%
randn('state',0)             % Start from a known state.
x = randn(1000,1);             % 1000 Gaussian deviates ~ N(0,1).
y = filter(1,[1 -0.6 0.08],x);  % Create a stationary AR(2)
[PartialACF, Lags, Bounds] = parcorr(y , [] , 2)

PartialACF =

1.0000
0.5570
-0.0931
0.0249
-0.0180
-0.0099
0.0483
0.0058
0.0354
0.0623
0.0052
-0.0109
0.0421
-0.0086
-0.0324
0.0482
0.0008
-0.0192
0.0348
-0.0320
0.0062

Lags =

   0
   1
   2
   3
   4
   5
   6
   7
   8
   9
10
11
12
13
14
15
16
17
18
19
20

Bounds =

0.0633
-0.0633

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

2012-12-7 10:45:27

epoh 发表于 2012-11-25 09:18
lagmatrix.m 有兩個
一個在ucsd_garch toolbox
另一個在 James P. LeSage Econometrics Toolbox, 你要 ...

http://www.spatial-econometrics.com/arfima/

epoh老师，您好！
为什么在http://www.spatial-econometrics.com的主页上看不到/arfima
这个子目录？
输入http://www.spatial-econometrics.com/arfima/这个可以看到？
怎么看到和了解http://www.spatial-econometrics.com/根目录下还有其他像arfima这样的目录呢？

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

2012-12-7 10:52:02

epoh老师，您好！
如何用附件中的包用simulvar.m和calcvar.m
调用数据dati1.mat做VaR分析？
非常感谢！

附件列表

zhtvarm.zip

大小:272.7 KB

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

2012-12-7 15:38:25

zhangtao 发表于 2012-12-7 10:52
epoh老师，您好！
如何用附件中的包用simulvar.m和calcvar.m
调用数据dati1.mat做VaR分析？

calcvar.m  - terminal interface for VaR calculation
%VaR estimation.
deltavar       - delta-normal VaR.
simulvar       - hystorical simulation VaR.
montevar    - Monte Carlo VaR.
%in matlab command window
>>calcvar

%       Load data ------------------------------------------------------------------------

file=input('Which data? (1, 2, 3) '); 1
   if file==1 load('dati1');
      elseif file==2 load('dati2');
      else load('dati3');
end
%       Type of return --------------------------------------------------------------------

rendimento=input('Which return? (0 normal, 1 log) '); 0

%       Decay factor ----------------------------------------------------------------------

lambda=input('Decay factor? (x<=1) '); 0.9

%       Parameters ------------------------------------------------------------------------

t=input('Holding period? '); 1
p=input('Probability? '); 0.5

   %VaR method ------------------------------------------------------------------------

a=input('VaR methods? (1 delta-normal, 2 historical simulation, 3 Monte Carlo, 4 average) ');2

tipo=input('Mapping? (0 from risk factors, 1 from prices) '); 0
[svar] = simulvar(DR,RC,X,delta,valuta,legame,vm,t,p,tipo,rendimento)

svar =
0 from risk factors :  -0.0293
1 from prices    :  -0.0148

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

2012-12-7 21:22:38

epoh 发表于 2012-12-7 15:38
calcvar.m - terminal interface for VaR calculation
%VaR estimation.
deltavar - del ...

epoh老师，您好！
非常感谢您！我的意思是，如何用附件中的包用simulvar.m和calcvar.m
调用数据dati1.mat做VaR分析？就是可以套上自己的数据用simulvar.m和calcvar.m
计算VaR,而不是去执行这个主程序。您看simulvar.m和calcvar.m
如何才能套上自己的数据调用呢？
非常感谢！

基与蒙特卡罗模拟法的风险价值_VaR_及其在中国股票市场中的运用.zip
大小:(1.82 MB)

本附件包括：

基与蒙特卡罗模拟法的风险价值_VaR_及其在中国股票市场中的运用.nh

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

2012-12-8 10:45:53

epoh老师，您好！
您看看这个dynare程序什么地方有错？非常感谢！
??? Error: File: D:\MATLAB7\work\Untitled.m Line: 33 Column: 4
The expression to the left of the equals sign is not a valid target for an assignment.
??? Error: File: D:\MATLAB7\work\Untitled.m Line: 33 Column: 4
The expression to the left of the equals sign is not a valid target for an assignment.
??? Error: File: D:\MATLAB7\work\Untitled.m Line: 35 Column: 4
The expression to the left of the equals sign is not a valid target for an assignment.
??? Error: File: D:\MATLAB7\work\Untitled.m Line: 37 Column: 9
The expression to the left of the equals sign is not a valid target for an assignment.
??? Error: File: D:\MATLAB7\work\Untitled.m Line: 37 Column: 9
The expression to the left of the equals sign is not a valid target for an assignment.
??? Error: File: D:\MATLAB7\work\Untitled.m Line: 44 Column: 8
The expression to the left of the equals sign is not a valid target for an assignment.
??? Error: File: D:\MATLAB7\work\Untitled.m Line: 44 Column: 8
The expression to the left of the equals sign is not a valid target for an assignment.
??? Error: File: D:\MATLAB7\work\Untitled.m Line: 44 Column: 8
The expression to the left of the equals sign is not a valid target for an assignment.
>>

附件列表

dynaret.txt

大小:792 Bytes

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

2012-12-8 18:11:08

zhangtao 发表于 2012-12-7 21:22
epoh老师，您好！
非常感谢您！我的意思是，如何用附件中的包用simulvar.m和calcvar.m
调用数据dati ...

calcvar.m是个script
deltavar.m,simulvar.m,montevar.m才是用来计算VaR的function

你必须给出数据X,RF,RC:

dati1.xls
大小:(151 KB)

% X is the matrix of asset prices,
% RF the matrix of risk factors prices,
% RC the matrix of exchange rates
% The vector legame set the relations between assets and risk factors.
% Given 2 risk factors and 4 assets the vector (1 1 1 2) means that the
% first three assets are related with the first risk factors, the fourth
% asset with the second risk factor.
% legame=[1;1;1;1;2;3]
% vector vm of portfolio value.
% vm=[10;20;5;10;5;10]
% The vector valuta set the relation between tha asset and the exchange rate in the same manner of the
% vector legame, with the convention that 0 is for the referred exchange rate.
% valuta=[0;0;0;0;1;0]
% VaR estimation.
% deltavar       - delta-normal VaR.
% simulvar       - hystorical simulation VaR.
% montevar       - Monte Carlo VaR.
%load data dati1.xls
dati1=xlsread('dati1.xls','A2:J1046');
X=dati1(:,1:6);
RF=dati1(:,7:9);
RC=dati1(:,10);
legame=[1;1;1;1;2;3]
vm=[10;20;5;10;5;10]
valuta=[0;0;0;0;1;0]
%Type of return
rendimento=0  %(0 normal, 1 log)
[DX,DR,cambi,fattori,prezzi] = rendimenti(X,RF,RC,rendimento);

%Decay factor
lambda=0.9
if lambda==1
      [delta,dev,cor,C] = semplicecorr(DX,DR,legame);
      else [delta,dev,cor,C] = ewmacorr(DX,DR,legame,lambda);
end
%       Parameters ---------------------------------------------------------------------------------

t=1
p=0.5
% VaR method

% historical simulation
tipo=0  %(0 from risk factors, 1 from prices)

[svar] = simulvar(DR,RC,X,delta,valuta,legame,vm,t,p,tipo,rendimento) % -0.0293

% Monte Carlo
modello=0  %(0 linear, 1 log)
l=10000  %Number of simulations
[mvar] = montevar(delta,C,legame,valuta,cambi,fattori,prezzi,vm,t,p,modello,l,rendimento)  %  -0.0150

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝