search

JustinMShea / ExpectedReturns

40 stars 18 forks source link

Vignette: Value-oriented equity selection: The Devil in HML's Details #13

Closed JustinMShea closed 2 years ago

JustinMShea commented 4 years ago

Value factors to compare with HML

The Devil in HML’s Details (Asness and Frazzini, 2013)

Data

https://github.com/JustinMShea/ExpectedReturns/blob/master/inst/parsers/Devil-in-HML-Details.R

Background

_Investment Performance of Common Stocks in Relation to Their Price-Earnings Ratios: A Test of the Efficient Market (S. Basu, 1977)_

_The Cross-Section of Expected Stock Returns (Fama, French, 1992)_

Expected Returns (Ch 12), Engle (Ch 10).

JustinMShea commented 4 years ago

I've added a brief markdown template with sections to help guide you. A data parser has been started to get the factors from AQR. Copy/pasta the code into the vignette, and build associated transformations, and merge the factors against the individual equity samples found in FactorAnalytics, similar to what you did in the first vignette. Ping us with questions.

JustinMShea commented 4 years ago

@vi-to In addition to the code comment I made under commit 1e75089, break up your code into chunks, similar to what I did in #4 . Break one out for data loading and processing, fitFfm, and finally plm. Then, you and we, can more easily isolate errors in each section.

erolbicero commented 4 years ago

@JustinMShea @vi-to I finished editing the parsing script, the file now knits (at the time of this writing), however the build still fails for me.

jaymon0703 commented 4 years ago

I believe this issue is done. We have a working example of the HML Devil replication, and any additional work required to get the vignette into a "publishable" state (not that that is the objective i dont think) can be added to this issue once it is re-opened, or a new issue opened. Hope that is suitable to everyone.

JustinMShea commented 4 years ago

We need to transform this into a static vignette as suggested in issue #20. As part of this, in the data section, we should not source parsers, but fully display the parser code within the vignette. This will allow users to reproduce everything from this single vignette.

JustinMShea commented 3 years ago

Refactor into FactorAnalytics, and cleanup code comments