This thesis analyzes the timing issue related to informed trades under two different market frameworks. Firstly, the issue under the competitive market framework is analyzed. In financial market, a widely accepted assumption that competitive informed traders elect to trade immediately upon receiving their private information is now questioned. We propose a competitive rational expectations model to demonstrate that under some situations informed traders tend to trade late on their information because of the fear of adverse effect on prices from their informed trades. This phenomenon from delayed informed trades leads to the following: Price volatility will increase and adjacent price changes may exhibit positive serial correlation. Secondly, we turn to the alternative framework, i.e., a market microstructure framework. A large number of market microstructure models had already investigated the timing issue of informed trades. Most of them found that the competition among informed traders will make the informed traders incline to trade early than late on their information and the market therefore becomes more efficient. We develop a market microstructure model with competitive, risk averse informed traders and uninformed market makers. It is found that when the mass of informed traders is larger or the precision of private information is higher, the market becomes less efficient, that is, prices will delay revealing the private information and the market will postpone becoming liquid. Our results stand in contrast to those of other market microstructure models simply because informed traders in our model choose to trade late on their information. In conclusion, the thesis has proved that the timing to trade on their information is an important consideration for informed traders to determine their trading strategy maximizing their expected utility. Since this issue is seldom discussed in the previous literature, I deeply believe that there are many works to be done followed this work.