A State Space Framework for the Residual Income Valuation Model of Stock Prices

Abstract

We assess the empirical implications of the valuation model for equity prices developed in Ohlson (1995), by accounting for residual income information dynamics. A key assumption of the Ohlson (1995) residual income model stipulates that next period t + 1 residual income is a linear function of current period t residual income and a latent variable referred to as ‘other information’. This ‘other information’, assumed known in the current period t, contains information on next period t+1 abnormal earnings not reflected in current period t abnormal earnings. Previous literature has proxied this ‘other information’ variable with consensus analysts’ forecasts of earnings. In this study, we propose to estimate this latent ‘other information’ variable using a state space framework. Our method obviates the need for analysts’ earnings forecasts. We estimate the valuation model, within the embedded state space framework, using the Kalman filter recursive procedure. We estimate the model across a sample of stocks in the Dow Jones 30 and S&P 500 indices. We compare the model performance to a benchmark two-step regression approach used in previous work. Performance yardsticks indicate that our state space estimation approach shows promise in valuing stocks and predicting next period t+1 residual income relative to the benchmark approach.