Market neutral statistical arbitrage strategy by factor models in Tehran stock exchange

forecasting price movements is a challenging issue. So different statistical arbitrage strategies are devised to trade in exchanges. Some of these strategies are market neutral. Market neutral strategies are neutral to market movements and make profits in any situation. These strategies use long and short positions at the same time and this makes them unusable in exchanges like Tehran stock exchange that only long position is available. Purpose of this paper is devising a market neutral statistical arbitrage strategy which can be used in Tehran stock exchange. In devising this strategy we use principal component analysis to estimate market movements and calculate stocks idiosyncratic movements. To forecasting stocks idiosyncratic movements, which have mean reverting properties, we use Ornstein-uhlenbeck model. The strategy made 35% average annual return by considering transaction cost which is more than Tehran exchange index in the study period. results show this method is a good framework for devising statistical arbitrage strategies.