yangkedc1984 Goto Github PK

-deep_learning_in_asset_pricing

access-r-source

How to get at R source. I am sick of Googling this. I am writing it down this time.

adaptive_penalized_macro_factors

advance-simulation-methods-project-for-stock-price-forecasting

In general, the simulation is in a Geometric Brownian Motion framework, and with some modifications. I first visualize price and daily returns to find some trend and features of that stock, then I use the analytical insights to construct the model that can capture as much information as possible. Finally, I quantify the information, and put them into my simulation model.

advanced-deep-learning-with-keras

Advanced Deep Learning with Keras, published by Packt

advanced-macroeconometrics

advbayeslearncourse

Course material for the PhD course in Advanced Bayesian Learning

adventures-in-ml-code

This repository holds all the code for the site http://www.adventuresinmachinelearning.com

all-about-time-series-analysis

[Advanced] Practical Statistical Inference and Machine/Deep Learning for Time Series Forecasting

an-alternative-approach-to-forecast-the-volatility-of-multiscale-and-high-dimensional-market-data

Traditional methods for volatility forecast of multiscale and high-dimensional data like foreign-exchange and stock market volatility have both advantages and disadvantages which have been identified. In my project, I apply the Support Vector Machine (SVM) as a complimentary volatility method which is capable dealing of such type of data. SVM-based models may extract extra information of time series data and handle the long memory effect very well. Our Support Vector Machine for Regression (SVR) model has better result than the common GARCH (1, 1) model. The predictions are closer to the historical data and the error is lower. In addition, I test different kernels to see the performance difference. For my data, rbf kernel has an overall better performance than linear and polynomial kernels. I conclude that SVM-based model may be applied more frequently in the emerging field of high-frequency finance and in multivariate models for portfolio risk management.

application-of-transfer-entropy-to-assess-causality-in-metabolomic-networks

My third research project at the Centre for Computation, Mathematics and Physics in the Life Sciences and Experimental Biology at University College London (UCL CoMPLEX).

applied-methods-phd

Repo for Yale Applied Empirical Methods PHD Course

applied_metrics

A PhD course in Applied Econometrics and Panel Data

ar-dcc_residual_analysis

autoregressive-dynamic conditional correlation model with residual analysis, Designed for fMRI analysis(Dynamic Functional Connectivity), useful for other applications too.

artificial_neural_networks

A collection of Methods and Models for various architectures of Artificial Neural Networks

ash-orig

ashr

An R package for adaptive shrinkage

autoencoder_factor

simple implementation of Autoencoder Asset Pricing Models

awesome-network-analysis

A curated list of awesome network analysis resources.

b.a.-thesis

Codes used in the B.A. Thesis

balanced_quantile_predictive_regression

bayescomb

Data and code for paper "A Bayesian approach to combining multiple information sources"

bayesian-ms-var

This repository contains a matlab code to estimate a Bayesian Markov-Switching Vector Auto Regression Model

bayesian_ovfit_mixedscale_compositionaldata_mixturemodels

This repository contains functions for obtaining posterior samples of allocation variables in multiple Bayesian over-fitted (sparse finite) mixed-scale mixture models. Mixture models included: 1) Bayesian Tensor Mixture of Product Kernels model (BayesTMPK), 2) Modularized Tensor Factorizations (MOTEF), 3) Bayesian Mixture of Product Kernels (BayesMPK), 4) Bayesian Mixture of Multivariate Gaussians (BayesMixMultGauss). Functions 1 and 2 include ability to model compositional data with essential zeros. Functions 3 and 4 include ability to model non-zero compositional data.

bayesianms-var-gc

Bayesian Estimation of Markov-Switching VARs for Granger Causal Inference in R

bayeslearncourse

Bayesian Learning course at Stockholm University

bayesmar

Bayesian Median Autoregressive model for time series forecasting

bayesvar_tvp

R/C++ implementation of Bayes VAR models

bear-toolbox

The Bayesian Estimation, Analysis and Regression toolbox (BEAR) is a comprehensive (Bayesian Panel) VAR toolbox for forecasting and policy analysis.

bgvar

Toolbox for the estimation of Bayesian Global Vector Autoregressions in R.