2. MTECリサーチアーカイブ
  3. 論文/ワーキングペーパー
  4. 論文/ワーキングペーパー(2016年)



著者 霧生 拓也(MTEC)、枇々木規雄(慶應義塾大学)
タイトル Estimating Forward Looking Distribution with the Ross Recovery Theorem, Working Paper

Recently, Ross [2015] introduced a remarkable theorem, named the “Recovery Theorem.” It enables us to estimate the real world distribution from the risk neutral distribution derived from option prices under a particular assumption about a representative investor’s risk preferences. The real world distribution estimated with the Recovery Theorem is suitable for many financial problems such as market risk management andportfolio optimization due to its forward looking nature. However, it is not easy to derive the appropriate estimators because of an ill-posed problem in the estimation process. We propose a new method to derive the accurate solution by formulating the regularization term involving prior information. Previous studies propose methods to estimate the real world distribution, but they do not investigate the estimation accuracy. We conduct the numerical analysis with hypothetical data to show the effectiveness of the proposed method. We find the following three points from the results.

  1. Stabilizing the solution by introducing the regularization term is an effective method of accurately estimating a real world distribution from the Recovery Theorem.
  2. The proposed method can estimate a real world distribution more accurately than the Tikhonov method used by Audrino et al. [2015].
  3. Our criteria for selecting a regularization parameter can offer the appropriate parameter in most cases.
著者 川口 宗紀(MTEC)
タイトル Vine コピュラを用いた与信ポートフォリオの信用ストレステストモデル, 日本統計学会誌, 第45巻 (第2号), pp. 307-328
発表年月日/掲載誌 2016.3
著者 石部真人(MTEC)、角田康夫(青山学院大学非常勤講師(前期))、坂巻敏史(MUTB)
タイトル 「株式市場リターンの季節性と投資家のリスク回避傾向の季節変化」,証券アナリストジャーナル, Vol.54, No.2, Feb 2016, pp 68-77
発表年月日/掲載誌 2016.2