preview

The Test For Visa Inc. And Waters

Good Essays

In the ADF level test, we cannot reject the null hypothesis that there is a unit root in each series of both VISA INC. and WATERS. This is shown in the LEVEL columns above. The 1st difference columns above show the output of the ADF test for the first-differenced series. All the series are significant so that they reject the null hypothesis. We can say that the first-differenced series are I(0), so they are stationary. And we can say that each level series contains a unit root and is integrated of order 1(I(1)). In this part, I make the lag interval as 1 and do the Johansen’s Cointegration Test for 3 groups of data separately. This step is to test whether there is a long term relationship between those series so that we can decide whether …show more content…

The table above is the covariance matrix of the futures bid price and spot bid price of VISA INC. we can get σ_1^2, σ_12, and σ_2^2 to calculate HAS1 and HAS2. This table is from the VEC test of lnfutures_bid100 and lnfutures_ask100. We can get λ_1 and λ_2 to calculate HAS1 and HAS2. In conclusion, I get 3 groups of outputs of bid, ask and midpoint price for VISA INC. and WATERS separately. The following functions can help calculate the upper and lower bound of information shares. For VISA INC., when I use ask and bid prices, the average information shares of spot price is much larger than future prices so that the spot market contributes more to the price discovery process. However, when I use midpoint prices, they contribute roughly same to the price discovery process since both average information shares are about 0.5. For WATERS, the all 3 kinds of prices provide similar outputs. Their average information shares of future and spot prices are roughly 0.5. In conclusion, for VISA INC., the spot market contributes more for the price discovery process, but for WATERS, both SSF market and spot markets contribute roughly same to the price discovery process. However, it does not mean that the SSF market is not important. Contributing less than 50% does not mean that the SSF market is not important. Fung and Tse (2007) support this point. They suggest that the ratio can only indicate the difference of reaction speed. In my

Get Access