Maoning Wang 2016-06-27
来自cslt Wiki
last week:
1.test how to find 'k' in equation(3.4) in [LoMacKinlay1990] [1]
2.parameter estimation of the assumptions in LiBin's articles
3.learn to use ntstool in matlab to predict the next price
this week:
1.test other datasets on 'k' in equation(3.4)
2.find the reason why OLMAR works better than PAMR. My tools:(1)theoretical analysis on Perceptron (2) the relationship between the form of PAMR/OLMAR and equation(3.3) in [LoMacKinlay1990]