 |
 |
 |
 |
 |
 |
|
 |

 |
|
Áú¹® ]
ÃÍ·¹½ºÅ° ºÐÇØ¿Í ÃÍ·¹½ºÅ° ÆåÆ®°¡ °¢°¢ ¾î¶² Àǹ̸¦ Áö´Ï°í ÀÖ³ª¿ä? Áï °á±¹ ¾î¶² ¿ªÇÒÀ» ¼öÇàÇÏ´Â °ÍÀÎÁö, ¸ñÀûÀÌ ¹ºÁö ¼³¸í Á» ºÎʵ右´Ï´Ù. Ã¥À» ºÁµµ °³³äÀÌ È®½ÇÈ÷ Á¤¸³ÀÌ ¾ÈµÇ³×¿ä
[ ´äº¯ ]
FactorizationÀº ÀμöºÐÇØ¶ó´Â ¶æÀ̱¸¿ä, decompositionÀ̶ó°íµµ ÇÕ´Ï´Ù. ±×·¯´Ï±î Cholesky Factorizationµµ ¹«¾ð°¡¸¦ ÀμöºÐÇØÇÏ´Â ¹æ¹ýÀä, ±× '¹«¾ð°¡'¿¡ ÇØ´çÇÏ´Â °ÍÀÌ MatrixÀÔ´Ï´Ù. µû¶ó¼ Cholesky FactorizationÀº Matrix¸¦ ÀμöºÐÇØÇÏ´Â °ÍÀε¥, ±×³É 'Matrix'¿¡ root(Á¦°ö±Ù)¸¦ ¾º¿î °á°ú(¹°·Ð Matrix¿¡ root¸¦ ¾º¿ì¸é ±× °á°úµµ MatrixÀÇ ÇüŰ¡ µË´Ï´Ù.)¶ó°í »ý°¢ÇÏ½Ã¸é ½±½À´Ï´Ù.
¿¹¸¦ µé¾î º¼±î¿ä?
'2'¸¦ Çѹø ºÐÇØÇØº¸ÁÒ. '2'¸¦ µ¿ÀÏÇÑ ¼ýÀÚ(Factor)ÀÇ °öÀ¸·Î ºÐÇØÇϸé 2=root(2)*root(2)·Î ³ªÅ¸³¾ ¼ö ÀÖ½À´Ï´Ù. ¸¶Âù°¡Áö·Î Matrixµµ ±×¿Í °°Àº °öÀ¸·Î ³ªÅ¸³¾ ¼ö Àִµ¥ Matrix¸¦ µÎ Matrix(Factor)ÀÇ °öÀ¸·Î ºÐÇØÇÏ°Ô µÇ¸é Matrix¸¦ °öÇÏ´Â ¹æ¹ýÀÌ ÀϹÝÀûÀ¸·Î ¼ýÀÚ¸¦ ¼·Î °öÇÏ´Â ¹æ½Ä°ú´Â ´Ù¸£¹Ç·Î,
Original Matix=Lower triangular matrix(Matrix ¸ð¾çÀÌ ´ë°¢¼±À» Áß½ÉÀ¸·Î ¿ÞÂÊ ¾Æ·¡¿¡¸¸ ¼ýÀÚ°¡ ÀÖ°í ´ë°¢¼± À§ÂÊÀº ¿ø¼Ò°¡ ¸ðµÎ 0ÀÎ Çà·Ä) * Upper triangular matrix(Lower triangular matrixÀÇ ¿°ú ÇàÀ» ¹Ù²Û ÀüÄ¡ Çà·Ä)·Î ºÐÇØÇÒ ¼ö ÀÖ½À´Ï´Ù.
¸®½ºÅ©°ü¸®¿¡¼ Cholesky FactorizationÀº Monte Carlo Simulation¿¡¼ Random number¸¦ ÃßÃâÇÒ ¶§, ÃßÃâµÈ random number(randomÇÏ´Ï±î °¢ random numberµé°£ÀÇ correlationÀº 0À̰ÚÁÒ?)°£ÀÇ correlationÀÌ 0ÀÌ ¾Æ´Ï¶ó '¿øÇÏ´Â correlation'À» °¡Áöµµ·Ï random number¸¦ ¼öÁ¤ÇÒ ¶§ »ç¿ëÇÕ´Ï´Ù.
´ë·«ÀûÀ¸·Î ¸»¾¸µå¸®¸é ÃßÃâµÈ random numberµé(µû¶ó¼ Correlation=0)ÀÇ matrix¿¡ À§ Original matrix(risk factorµé°£ÀÇ ½ÇÁ¦ correlation matrix)¸¦ ºÐÇØÇÏ¿© ³ª¿Â Lower triangular matrix¸¦ °öÇϸé, ¼·Î°£ÀÇ correlationÀÌ 0ÀÎ random numberµéÀ» ¿øÇÏ´Â correlation±¸Á¶¸¦ °¡Áö´Â random numberµé·Î ¹Ù²Ü¼ö ÀÖ½À´Ï´Ù.
[Ãâó] ÃÍ·¹½ºÅ° ÀμöºÐÇØ|ÀÛ¼ºÀÚ ÀÓäâ
======================================================
´ÙÀ½Àº À̸¦ ÀÌ¿ëÇÑ ¼¼ °³ÀÇ Àڻ갡°Ý ½Ã¹Ä·¹ÀÌ¼Ç °á°úÀÌ´Ù. ÀÚ»êA¿Í BÀÇ »ó°ü°è¼ö¸¦ 0.9·Î ÁØ °á°ú Â÷Æ®¿¡¼µµ µÎ ÀÚ»êÀº ¸Å¿ì Ä£ÇÏ°Ô ¿òÁ÷À̰í ÀÖ´Ù.
³µ¥¾øÀÌ Â÷Æ®¸¦ º¸¿©ÁÖ¸ç »ó°ü°è¼ö°¡ ¾î¼°í Àú¼°í ÇÏ´Â °ÍÀº ÃÍ·¹½ºÅ° ºÐÇØ¸¦ ÀÌ¿ëÇÑ ¸óÅ× Ä«¸¦·Î ½Ã¹Ä·¹À̼ÇÀ» ²ôÁý¾î ³»·Á´Â °ÍÀÌ´Ù. Cholesky Decomposition ¶Ç´Â Cholesky FactorizationÀº »ó°ü°ü°è¸¦ °¡Áö´Â 2°³ ÀÌ»óÀÇ ³¼ö¸¦ ¸¸µé°íÀÚ ÇÒ¶§ ÀÌ¿ëÇÑ´Ù. ¸óÅ×Ä«¸¦·Î ½Ã¹Ä·¹À̼ǿ¡ ´ëÇÑ ±ÛÀ» º¸¸é ´ë°³ ÇϳªÀÇ ³¼ö¸¦ ¹Ú½º¹Ä·¯¹ý µîµîÀ¸·Î »ý¼ºÇÏ¿© ½Ã¹Ä·¹À̼ÇÀ» ÇÏ´Â °ÍÀÌ ´ë´Ù¼öÀÌ´Ù. ±×·¯³ª ELS¿Í °°Àº multi-assetÀÇ °æ¿ì Àڻ갣ÀÇ »ó°ü°ü°è¸¦ °í·ÁÇÏ¿© ³¼ö¸¦ »Ì¾Æ³»¾î¾ß ÇÑ´Ù. Àڻ갣¿¡ »ó°ü°ü°è°¡ 0À̶ó¸é ¸ð¸¦±î...? ¾Ïư Cholesky Decomposition¿¡ ´ëÇÑ Á¤ÀǸ¦ º¸¸é Given a symmetric positive definite matrix A, the Cholesky decomposition is an upper triangular matrix U such that A = transpose(U)U.
A=UTU
´ÙÀ½ÀÇ ÄÚµå´Â »ó°ü°è¼öÇà·ÄÀ» ¸Å°³º¯¼ö·Î ÃÍ·¹½ºÅ°ºÐÇØ¸¦ ÇÏ´Â vbnum.com¿¡¼ °¡Á®¿Â °ÍÀÌ´Ù.
Function Cholesky(Mat As Range) Dim A, L() As Double, s As Double A = Mat n = Mat.Rows.Count M = Mat.Columns.Count If n <> M Then Cholesky = "?" Exit Function End If ReDim L(1 To n, 1 To n) For j = 1 To n s = 0 For k = 1 To j - 1 s = s + L(j, k) ^ 2 Next k L(j, j) = A(j, j) - s If L(j, j) <= 0 Then Exit For L(j, j) = Sqr(L(j, j)) For i = j + 1 To n s = 0 For k = 1 To j - 1 s = s + L(i, k) * L(j, k) Next k L(i, j) = (A(i, j) - s) / L(j, j) Next i Next j Cholesky = LEnd Function³¼ö»ý¼ºÀº...
|
|
|
copyright ¨Ï 2005 ZIP365.COM All rights reserved |
|