In Japanese Trade Statistics 1, we find that China is the driving force behind the increase of the Japanese trade with Asian countries from 1988 to 2015.
Following this, we want to know what kinds of products contribute most to this change of trading pattern.
To answer this question, we first have a look at the structure of commodities in the Japanese Trade.
Continue reading “Japanese Trade Statistics Analysis 2”
Back to 1988, The North America is the biggest trading partner of Japan, However, after 1990, Asia start to replace the position of North America and lead with a very huge gap compared to other areas.
Continue reading “Japanese Trade Statistics Analysis 1”
What is hinge loss
The hinge loss is a loss function used to train the machine learning classifier, which is
where = -1 or 1 indicating two classes and represents the output from our classifier.
However, the SVM I know is like
So what is the relation between the two? Are they just two perspectives to look at the same model?
Continue reading “Derive hinge loss from SVM”
The logistic regression model is a simple but popular generalized linear model. It is used to make classification on binary or multiple classes. Here, we will try to implement this model with python, test the results on simulated data and compare its performance with the logistic regression module of scikit-learn.
Continue reading “Logistic regression with softmax function”