用MATLAB进行蒙特卡洛模拟？

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

请问有谁知道用MATLAT如何进行蒙特卡洛模拟？我要做毕业论文，处理上证指数。跪求，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

分享0 收藏6 回帖

关键词：MATLAB 蒙特卡洛模拟 matla atlab 蒙特卡洛 MATLAB 模拟 蒙特卡洛

蒙特卡洛模拟?用R语言吧，比较方便，也不怎么需要编程

如果连这个都不会，还是别做了。

请不要灌水啊，哈哈。不会才问吗！

不知道你是问哪一方面的。
在最新版本的matlab里，econometrics toolbox 里的 期权定价，信用风险分析等，这些demo都是蒙特靠罗模拟的。

我贴一个我去年改装的GARCH 的蒙特卡洛模拟的程序吧。数据应该是调用demo里的。安装了matlab，就有那个数据的。
可以直接调用。

哦，处理上证指数哦。哈哈。

conn=yahoo;
enddate=datestr(today);
SZindex=fetch（‘000001.ss’,'close','01/01/2001',enddate,d);
忘记最后一个参数了，好像设置为每一日的数据，应该有单引号，也好像没有，可以 help fetch。

1. conn=yahoo;
   
2. SZindexprice=fetch(conn,'000001.ss','close','01/01/2001',05/23/2011',d);
   
3. retSZindex=price2ret(SZindexprice);
   
4. markpound = retSZindex;
   
5. [H,pValue,Stat,CriticalValue] = lbqtest(markpound-mean(markpound),[10 15 20]',0.05);
   
6. [H1,pValue,Stat,CriticalValue] = lbqtest((markpound-mean(markpound)).^2,[10 15 20]',0.05);
   
7. [coeff,errors,LLF,innovations,sigmas,summary] = garchfit(markpound);
   
8. spec = garchset('VarianceModel','GJR','R',1,'M',1,'P',1,'Q',1);
   
9. spec = garchset(spec,'Display','off','Distribution','T');
   
10. [coeff,errros,LLF,eFit,sFit] = garchfit(spec,markpound);
    
11. [eFit,sFit] = garchinfer(coeff,markpound);
    
12. 
    
13. spec1=garchset;
    
14. [coeff1,erros1,LLF1,eFit1,sFit1]=garchfit(spec1,markpound);
    
15. 
    
16. 
    
17. horizon = 20;
    
18. nPaths = 20000;
    
19. strm = RandStream('mt19937ar','Seed',5489);
    
20. RandStream.setDefaultStream(strm);
    
21. [eSim,sSim,ySim] = garchsim(coeff,horizon,nPaths,[],[],[],eFit,sFit,markpound);
    
22. Y=ret2price(ySim);
    
23. [eSim1,sSim1,ySim1]=garchsim(coeff1,horizon,nPaths,[],[],[],eFit1,sFit1,markpound);
    
24. Y1=ret2price(ySim1);
    
25. 
    
26. subplot(2,1,1),plot(Y)
    
27. subplot(2,1,2),plot(Y1)
    
28. 
    
29. %% expirement hist plot usd prctile
    
30. exret=mean(Y1)*100;

复制代码

1. %% This program is writed by Danniel tulip lew
   
2. %   Danny have this program`s copyright  @2010
   
3. 
   
4. %% The first step:Loading data to whcih will be used to analysis
   
5. CleanUp;
   
6. 
   
7. load mygarchdata
   
8. 
   
9. markpound = SSret;
   
10. figure
    
11. plot(markpound)
    
12. title('市场序列')
    
13. 
    
14. adfval = adftest(SSret);
    
15. disp('The test value of Augmented Dickey-Fuller test for unit root');
    
16. disp('------------------------------------------------------------');
    
17. disp(adfval)
    
18. 
    
19. kpssval = kpsstest(SSret);
    
20. disp('The test value of kpsstes');
    
21. disp('---------------------------');
    
22. disp(kpssval)
    
23. 
    
24. [h,pValue,stat,cValue] = archtest(SSret,1:5,0.01);
    
25. disp('Engle test for residual heteroscedasticity');
    
26. disp('---------------------------------------------');
    
27. disp(h)
    
28. disp('----------------------------------------------');
    
29. disp(pValue);
    
30. 
    
31. %% Bulding GARCH Model and fit the parameters
    
32. spec = garchset('VarianceModel','GJR','R',1,'M',1,'P',1,'Q',1);
    
33. spec = garchset(spec,'Display','off','Distribution','T');
    
34. [coeff,errros,LLF,~,~] = garchfit(spec,markpound);
    
35. [eFit,sFit] = garchinfer(coeff,markpound);
    
36. 
    
37. spec11=garchset;
    
38. spec11=garchset(spec11,'Display','off');
    
39. [coeff1,erros1,LLF1,eFit1,sFit1]=garchfit(spec11,markpound);
    
40. 
    
41. %% Monte carlo simulation and visulise the simulation result using
    
42. horizon = 30;
    
43. nPaths = 10000;
    
44. strm = RandStream('mt19937ar','Seed',5489);
    
45. RandStream.setDefaultStream(strm);
    
46. [eSim,sSim,ySim] = garchsim(coeff,horizon,nPaths,[],[],[],eFit,sFit,markpound);
    
47. Y=ret2tick(ySim);
    
48. [eSim1,sSim1,ySim1]=garchsim(coeff1,horizon,nPaths,[],[],[],eFit1,sFit1,markpound);
    
49. Y1=ret2tick(ySim1);
    
50. % plot result
    
51. figure
    
52. subplot(2,1,1),plot(Y)
    
53. subplot(2,1,2),plot(Y1)
    
54. 
    
55. %% Garch simulation and garch predict  
    
56. NumPeriods=30;
    
57. fprintf('Garchpred result: Sigma and Mena\n');
    
58. [SigmaForecast,MeanForecast,sigmaTotal,meanRMSE] = ...
    
59. garchpred(coeff,markpound,NumPeriods);
    
60. disp('Sigma')
    
61. disp(SigmaForecast)
    
62. disp('Mean')
    
63. disp(MeanForecast)
    
64.    
    
65. 
    
66. %% Standardized residuel
    
67. 
    
68. figure
    
69. autocorr(markpound);
    
70. % Compare Sigmas
    
71. figure
    
72. plot(SigmaForecast,'.-b')
    
73. hold('on')
    
74. grid('on')
    
75. plot(sqrt(mean(sSim.^2,2)),'.r')
    
76. title('Forecast of STD of Residuals')
    
77. legend('forecast results','simulation results')
    
78. xlabel('Forecast Period')
    
79. ylabel('Standard Deviation')
    
80. 
    
81. % Compare Returns
    
82. figure
    
83. plot(MeanForecast,'.-b')
    
84. hold('on')
    
85. grid('on')
    
86. plot(mean(ySim,2),'.r')
    
87. title('Forecast of Returns')
    
88. legend('forecast results','simulation results',4)
    
89. xlabel('Forecast Period')
    
90. ylabel('Return')
    
91. 
    
92. % Compare Standard Errors
    
93. figure
    
94. plot(meanRMSE,'.-b')
    
95. hold('on')
    
96. grid('on')
    
97. plot(std(ySim'),'.r')
    
98. title('Standard Error of Forecast of Returns')
    
99. legend('forecast results','simulation results')
    
100. xlabel('Forecast Period')
     
101. ylabel('Standard Deviation')
     
102. 
     
103. 
     
104. 
     
105. %% Hypothesis test: | lrationtest |
     
106. % A likelihood ratio test is another way to compare the model which is
     
107. % beest
     
108. [H,pValue]=lratiotest(LLF,LLF1,1);
     
109. fprintf('Result of the GARCH model compare using "lrationtest"\n');
     
110. if H==0
     
111.         disp('The GARCH(1,1) model is suported at the 5% significant level')
     
112.        else
     
113.         disp('The GJR(1,1) model is suported at the 5% significant level')
     
114. end
     
115. 
     
116. %-----------------------------EOF------------------------------------------

复制代码