It's nice to meet you.

It’s nice to meet you. My name is Ayato Ashihara, and I’m a first-year graduate of an asset management company in Tokyo.

I’d like to introduce myself briefly. I’m 24 years old and graduated from a graduate school. I majored in macroeconomics, especially DSGE models, state space models, Kalman filters, Bayesian estimation, and MCMC. I also did some natural language processing as research support for my advisor. I originally wanted to be an academic researcher, but due to financial problems, I had to find a job. Now I’m doing menial work in sales support.

I think this blog will be a memorandum of my research hobby as I cannot give up my interest in research.

I am currently working on the following two research projects.

1. horse betting version of the factor model
2. quarterly GDP projection model

The first question is whether it is possible to create a model for constructing a horse betting portfolio that aims for a collection rate of over 100%, from the perspective that the horse betting market is more speculatively attractive than the stock market. For stocks, there is a factor model like Pharma French, but I would like to see if we can apply it to create a horse-trading version of the factor model.

Secondly, we are aware of the problem of the low accuracy of the recent preliminary quarterly GDP report, so we wondered if it would be possible to construct a new, highly accurate forecasting model. In particular, I would like to use machine learning to create a model that incorporates data that is not normally used in macroeconomic research, rather than a model based on macroeconomic theory.

We’ve only just begun our research. I’ll do my best to keep you interested…

Best regards.