股價風險之評估-修正之 Ohlson 股權評價模型與幾何布朗運動之比較

Value-at-Risk Evaluation in Stock Price-A Comparison between Extended Ohlson Model and Geometric Brownian Motion Model

徐珮珊(Pei-Shan Hsue)

指導教授 ： 吳博欽

中原大學/商學院/國際貿易研究所/碩士(2008年)

https://doi.org/10.6840/CYCU.2008.00061

摘要

著名會計學者 Ohlson (1995) 利用當期財務報表中的帳面價值、異常盈餘，以及非會計資訊建立股價評價模式，惟 Ohson 並未對非會計資訊做出明確的說明與定義，故本文先以 Ohson 股權評價模型為基礎，將其中的非會計資訊以總體經濟變數與公司財務比率表示，進行股價評估與預測。其次，依所建立的修正後Ohlson評價模型，以蒙地卡羅模擬法進行股價模擬，並衡量公司股價之風險值。最後，進行樣本外預測，比較幾何布朗運動 (Geometric Brownian Motion) 與修正後 Ohlson 股權評價模型在預估股價風險上之績效差異。
實證上以臺指五十成份股作為研究對象，並採用追蹤資料 (Panel Data) 的隨機效果 (Random Effect Model) －固定斜率模型進行實證分析。為了進行實證研究與績效之評估，其樣本期間分為樣本內 ( 1999 年第四季至 2006 年第三季) 及樣本外 ( 2006 年第四季至 2007 年第三季) 兩部份。
根據實證結果，可獲得下列結論：
一、在未考量非會計資訊之原始 Ohlson 股權評價模型
下，其帳面價值及異常盈餘皆對股價有顯著正向影
響之結果。
二、加入以總體經濟變數及公司財務比率因子作為非會
計資訊後，重新評估 Ohlson 股權評價模型，皆能
有效提升模型之解釋能力。
三、整體而言，以修正後 Ohlson 股權評價模型估算出
的風險值皆低於幾何布朗運動所推估之風險值。此
外，由樣本外預測能力比較得知，修正後Ohlson股
權評價模型在評估風險值之績效上優於幾何布朗運
動；而股價之評估績效則以幾何布朗運動為佳。

關鍵字

Ohlson 股權評價模型 ； 追蹤資料 ； 蒙地卡羅模擬 ； 風險值

並列摘要

In his equity valuation model, Ohlson (1995) shows that stock price can be expressed as the summation of abnormal earnings, book value, and non-accounting information. However, numbers shown in financial statement (such as abnormal earnings and book value) can only provide a limited measure of stock price, because they do not reflect other important information about firms’ fundamentals. Especially, Ohlson did not define the non-accounting information very clearly, so this thesis first revises Ohlson equity valuation model by replacing his “non-accounting information” with several important factors, including “macroeconomic variables” and “company factors” and re-evalute the stock price. Secondly, based on the revised Ohlson equity evaluation model, I use Monte Carlo simulation to calculate Value-at-Risk (VaR) for corporate. Finally, to measure and compare the VaRs of the revised Ohlson model and Brownian Motion, then choose the model that can forecast stock price and VaR more officiently.
In empirical study, I use Random effect model with fixed slope to evaluate the stock price in Taiwan 50 index constituents. To differentiate between empirical studies and forecast perform evaluation, sample periods spans from 1999 Q4 to 2006 Q3 and 2006 Q4 to 2007 Q3 separately.
The empirical results show as follows:
1.Financial indices, including book value and abnormal profit in original Ohlson equity valuation model all have positive effect on stock price.
2.By adding macroeconomic variables and company factors to represent non-accounting information in original Ohlson equity valuation model, the fitness of the revised model is improved.
3.In most sample companies, the VaR calculated from revised Ohlson equity evaluation model is lower than that from Geometric Brownian movement. In addition, for out-of-sample forecast performance, the revised Ohlson equity evaluation model has better VaR forecast performance than Geometric Brownian movement. However, in stock price forecasting, the result is complete opposite.
Keywords:

並列關鍵字

Panel Data ； Value-at-Risk ； Ohlson equity valuation model ； Monte Carlo simulation

參考文獻

3. Bachelier, L. (1900). Th

一、英文部分

1. Abdullah and Hayworth (1993) , Macroeconometrics of stock price fluctuations, Journa1 of Business and Economics.

2. Abell, J. D. and T. M. Krueger (1989),“Macroeconomic influences on beta.” Journal of Economics and Business, 41, 185-193.