• ## Analysis of stock market volatility by GARCH model of R language time series

Time：2021-11-16

Original link:http://tecdat.cn/?p=22360  In this article, we will learn a standard method to establish volatility model in price series, namely generalized autoregressive conditional heteroscedasticity (GARCH) model. The idea of GARCH model of price fluctuation is to use the recent realization of error structure to predict the future realization of error structure. More simply, we often see […]

• ## Pre machine learning (I): mathematical symbols and Greek letters

Time：2021-10-7

This article is included inMachine learning pre tutorial series。 This paper lists the commonly used mathematical notations of machine learning, including algebra, calculus, linear algebra, probability theory, set theory, statistics and Greek letters. Algebra Symbol name describe example (f∘g) Compound function Nested function (f∘g)(x)=f(g(x)) ∆ Delta Change / difference ∆x=x_1-x_0 e Euler number e=2.718281828 \$ […]

• ## Python uses time-varying Markov regime switching autoregressive model to analyze economic time series

Time：2021-9-13

Original link:http://tecdat.cn/?p=22617 ============================================================== This paper provides an example of using Markov transformation model in statistical model to reproduce some results proposed by Kim and Nelson (1999). It applies the filter of Hamilton (1989) and the smoother of Kim (1994). %matplotlib inline import numpy as np import pandas as pd import statsmodels.api as sm from pandas_datareader.data import DataReader from datetime import datetime  DataReader(start=datetime(1947, 1, 1), end=datetime(2013, 4, 1))   Hamilton (1989) Markov Switching Model_  _ model_） This […]

• ## Hyperparametric Bayesian Optimization

Time：2021-9-7

Hyper parameters are parameters that define the model or the training process. They are relative to model parameters, such as the network structure of the target detection network, the selection of activation function, the size of learning rate, the size of anchor, etc. they all belong to hyper parameters. Hyper parameters have a great impact […]

• ## Learn the use of statistical extension functions in PHP

Time：2021-9-7

Friends who do statistical related systems must have learned concepts such as normal distribution, variance and standard deviation. In PHP, there are also corresponding extension functions specially developed for these statistical related functions. The stats extension library we are going to learn today is this kind of operation function. Of course, I haven’t done any […]

• ## R language multiplication GARCH model is used to predict the volatility of high-frequency trading data

Time：2021-8-28

Original link:http://tecdat.cn/?p=22692  Original source:Tuo end data tribal official account Over the past decade, interest in high-frequency trading and models has multiplied. Although I have some doubts about the effectiveness of signals in high-frequency noise, I decided to use GARCH model to study the statistical model of yield. Different from daily and lower frequency benefits, intra […]

• ## Data dimensionality reduction: principal component analysis

Time：2021-8-6

preface What is called principal component analysis? Let’s first look at a graph of an ellipse. If you were asked to find a line so that all points on the ellipse map the most scattered points on the line and retain the most information, how would you choose this line? In the following figure, horizontal […]

• ## What is heteroscedasticity

Time：2021-7-23

Today, let’s talk about heteroscedasticity. Before heteroscedasticity, let’s talk about another concept similar to heteroscedasticity: homovariance. What is homovariance? The same variance = the same + variance, as the name suggests, is the same variance. What is the variance? Variance is used to reflect the fluctuation of data. The same variance means that the fluctuation […]

• ## A case study of R language penalized logistic regression (Lasso, ridge regression) high dimensional variable selection classification model

Time：2021-6-9

Link to the original text:http://tecdat.cn/?p=21444 Logistic regression is a common method in research, which can screen influencing factors, predict probability and classify. For example, the data obtained by high pass sequencing technology in medical research challenges the selection of high-dimensional variables. Penalized logistic regression can select variables and estimate coefficients for high-dimensional data, and its […]

• ## Arma-egarch model of R language and integrated prediction algorithm are used to predict the actual volatility of SPX

Time：2021-5-7

Link to the original text:http://tecdat.cn/?p=12174 – introduce This paper compares several time series models to predict the daily growth of the SP500 indexReal volatility. The benchmark is arma-egarch model of SPX daily return series. It is compared with GARCH model   。 Finally, the ensemble prediction algorithm is proposed. Assumptions The actual volatility is invisible, […]

• ## Data dimension reduction: principal component analysis

Time：2021-4-19

preface What is called principal component analysis? Let’s first look at a graph of an ellipse. If you were asked to find a line so that all the points on the ellipse mapped on the line were the most scattered and the most information remained, how would you choose this line? In the figure below, […]

• ## Database transaction back test series 3: multi factor alpha strategy optimal factor weight

Time：2021-4-14

In the second part of this series（Multi factor alpha strategy backtesting）In this paper, we test the four quantitative factors of the U.S. stock market. Here, we’ll use the built-inquadprogFunction to optimize the mean variance of each factor weight to determine the best factor weight. After the script provided in the second part of this series […]