<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Checkmate9949的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Checkmate9949</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Checkmate9949]]></copyright><item><title><![CDATA[Spyder encounter an internal problem]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/152890993</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/152890993</guid><author>Checkmate9949</author><pubDate>Fri, 10 Oct 2025 09:12:46 +0800</pubDate><description><![CDATA[尝试打开dataframe和dict时报错。]]></description><category></category></item><item><title><![CDATA[记一次环境迁移过程]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/150224466</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/150224466</guid><author>Checkmate9949</author><pubDate>Mon, 11 Aug 2025 20:28:12 +0800</pubDate><description><![CDATA[现有的包已非常便捷，但环境配置、不同工具包的适配非常花时间，跑通一次代码后直接转移环境非常必要。将环境部署到目标机D:\llm-env：tar -xzf llm-env-win64.tar.gz -C D:\llm-env。cd到放置llm-env-win64.tar.gz的位置：cd C:\Users\你的用户名\Downloads。conda pack -n [环境机环境名] -o llm-env-win64.tar.gz。激活环境：conda activate D:\llm-env。]]></description><category></category></item><item><title><![CDATA[解决statsmodels版本冲突问题]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/150215566</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/150215566</guid><author>Checkmate9949</author><pubDate>Mon, 11 Aug 2025 14:47:52 +0800</pubDate><description><![CDATA[pip uninstall statsmodel -y  # 可能需要重复执行以确保完全卸载。]]></description><category></category></item><item><title><![CDATA[【Debug】Failed to get convolution algorithm. This is probably because cuDNN failed to initialize]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/130457230</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/130457230</guid><author>Checkmate9949</author><pubDate>Mon, 01 May 2023 17:07:21 +0800</pubDate><description><![CDATA[使用CUDNN-8.1.1，CUDA-11.2.2，Tensorflow-GPU-2.6.0，Python-3.9。使用PIP更新Tensorflow。运行全连接层正常，运行CNN出错。]]></description><category></category></item><item><title><![CDATA[read_excel() got an unexpected keyword argument ‘parse_date‘]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/122331262</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/122331262</guid><author>Checkmate9949</author><pubDate>Wed, 05 Jan 2022 20:20:29 +0800</pubDate><description><![CDATA[改为：parse_dates
]]></description><category></category></item><item><title><![CDATA[在Mathtype中打出双期望符号的两种方式]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/122251444</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/122251444</guid><author>Checkmate9949</author><pubDate>Fri, 31 Dec 2021 10:45:42 +0800</pubDate><description><![CDATA[1、直接在mathtype中复制：\mathbb{E}





2、三步走

（1）设置文本样式：



(2)选择Euclid Math Two



(3)打出E，设置样式为文本即可






]]></description><category></category></item><item><title><![CDATA[【Debug】unhashable type: ‘numpy.ndarray‘]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/122094833</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/122094833</guid><author>Checkmate9949</author><pubDate>Wed, 22 Dec 2021 20:57:51 +0800</pubDate><description><![CDATA[1、问题：

遍历出错


def _create_var_idx_associations(self, varList):
 idx2var = {idx:var for idx, var in enumerate(varList)}
 var2idx = {var:idx for idx, var in enumerate(varList)}
 return idx2var, var2idx

2、解决方法：

确保写入npz的permno的stock2格式：

(X,)

而不是(X...]]></description><category></category></item><item><title><![CDATA[Matlab triang()函数]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/121265106</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/121265106</guid><author>Checkmate9949</author><pubDate>Thu, 11 Nov 2021 11:57:00 +0800</pubDate><description><![CDATA[例子：W=59，B=triang(W)

n=30时，B刚好为1




]]></description><category></category></item><item><title><![CDATA[【西瓜书阅读笔记】06支持向量机]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/121013727</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/121013727</guid><author>Checkmate9949</author><pubDate>Thu, 28 Oct 2021 14:55:49 +0800</pubDate><description><![CDATA[零、Maximal margin classifier and Support Vector classifier



https://www.youtube.com/watch?v=efR1C6CvhmEhttps://www.youtube.com/watch?v=efR1C6CvhmE



一、间隔与支持向量：

1、通过升维寻找使margin最大的threshold



注意：该图表示三维空间，+对应y=1，-对应y=-1。



2、用线性方程描述超平面：法向量w和位移b决定了该...]]></description><category></category></item><item><title><![CDATA[【文献阅读】Do economic variables forecast commodity futures volatility]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120862467</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120862467</guid><author>Checkmate9949</author><pubDate>Wed, 20 Oct 2021 11:56:10 +0800</pubDate><description><![CDATA[1、The economic determinants of RV 波动率的经济决定因素：大宗商品供给的季节性、到期时间以及存货水平

Anderson and Danthine (1983) hypothesize that the key determinant of volatility is the time at which the production uncertainty is resolved.The uncertainty resolution is seasonal, for ins.]]></description><category></category></item><item><title><![CDATA[【西瓜书阅读笔记】05神经网络]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120816280</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120816280</guid><author>Checkmate9949</author><pubDate>Sun, 17 Oct 2021 21:30:33 +0800</pubDate><description><![CDATA[一、神经元Neuron





二、感知机Perceptron与多层网络


]]></description><category></category></item><item><title><![CDATA[【文献阅读】Journal of Finance76卷5期]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120795575</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120795575</guid><author>Checkmate9949</author><pubDate>Sat, 16 Oct 2021 11:45:11 +0800</pubDate><description><![CDATA[1、The Cross Section of MBS Returns抵押贷款支持证券(MBS)横截面收益

We present a simple, linear asset pricing model of the cross section of Mortgage-Backed Security (MBS) returns. MBS earn risk premia as compensation for their exposure to prepayment risk. We measure pr.]]></description><category></category></item><item><title><![CDATA[【文献阅读】Forecasting multivariate realized stock market volatility]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120669785</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120669785</guid><author>Checkmate9949</author><pubDate>Sat, 09 Oct 2021 14:57:23 +0800</pubDate><description><![CDATA[1、Introduction

The variances and covariances of stock returns vary over time (e.g. Andersen et al., 2005). As a result, many important financial applications require a model of the conditional covariance matrix. Three distinct categories of methods for es]]></description><category></category></item><item><title><![CDATA[【文献阅读】如何写好文献综述？]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120665865</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120665865</guid><author>Checkmate9949</author><pubDate>Sat, 09 Oct 2021 10:50:30 +0800</pubDate><description><![CDATA[转载自：

搞定文献综述写作，只需掌握这四个关键点就能盘它！ (qq.com)https://mp.weixin.qq.com/s/uKTGILntnktGajZ6e5cGtw



1、Coverage丨Analysis丨Interpretation

Coverage

围绕研究主题，对数据库以及相关引文进行全面细致的介绍。

Analysis

一篇好的文献综述不仅会告诉读者关于这个话题现有的研究及相关性，还会告诉读者它的意义、含义及局限性，为将今后的研究指向有益的方向。

将该研究方向抽丝剥茧，概括]]></description><category></category></item><item><title><![CDATA[【西瓜书阅读笔记】04决策树 第二部分]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120588526</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120588526</guid><author>Checkmate9949</author><pubDate>Sat, 02 Oct 2021 21:32:06 +0800</pubDate><description><![CDATA[一、CART算法回归树

1、原理：

根据阈值计算均方根误差之和，并选取均方差误差和最小的阈值：

防止过度拟合：设定最小分割样本数https://www.youtube.com/watch?v=g9c66TUylZ4https://www.youtube.com/watch?v=g9c66TUylZ4https://www.youtube.com/watch?v=g9c66TUylZ4









2、公式：



3、Sklearn是通过CART算法生成决策树



二、剪枝处理-预剪枝
...]]></description><category></category></item><item><title><![CDATA[【西瓜书阅读笔记】04决策树 第一部分]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120588376</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120588376</guid><author>Checkmate9949</author><pubDate>Sat, 02 Oct 2021 19:33:47 +0800</pubDate><description><![CDATA[第四章 决策树
1 基本流程

2 划分选择
随着划分过程不断进行，我们希望决策树的分支节点所包含的样本尽可能属于同一类别，即结点的“纯度”(purity)越来越高。
2.1 信息增益
2.1.1 什么是信息熵
https://www.zhihu.com/question/22178202
什么是熵：一种事物的不确定性。
信息：消除不确定性的事物。
信息的功能：调整概率；排除干扰；确定情况（比如卖瓜的人说了一句，包熟包甜）。
噪音：不能消除某人对某件事情不确定性的事物。
数据 = 信息 + 噪音
2.1.2]]></description><category></category></item><item><title><![CDATA[【文献阅读】The role of news sentiment in oil futures returns and volatility forecasting]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120583785</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120583785</guid><author>Checkmate9949</author><pubDate>Sat, 02 Oct 2021 11:12:42 +0800</pubDate><description><![CDATA[0、摘要

In this paper, we extract the qualitative information from crude oil news headlines, and develop a novel VMD- BiLSTM model with investor sentiment indicator for crude oil forecasting. 

本文中，我们提取了原油新闻标题的定性信息，develop a novel VMD-BiLSTM模型进行原油价格预测。

Firs]]></description><category></category></item><item><title><![CDATA[【代码实践】使用Garch模型估计VaR]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120568220</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120568220</guid><author>Checkmate9949</author><pubDate>Thu, 30 Sep 2021 15:59:55 +0800</pubDate><description><![CDATA[title: “Value at Risk estimation using GARCH model”
author: “Ionas Kelepouris &amp; Dimos Kelepouris”
date: “July 6, 2019”
output:
html_document:
fig_height: 7
fig_width: 10
highlight: tango
toc: yes
knitr::opts_chunk$set(echo = TRUE , warning = FALSE , e.]]></description><category></category></item><item><title><![CDATA[【高级微观经济学】利润最大化]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120517962</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120517962</guid><author>Checkmate9949</author><pubDate>Mon, 27 Sep 2021 22:20:21 +0800</pubDate><description><![CDATA[一、目标函数：利润最大化





二、利润最大化必要条件

1、内点解

（1）单要素情况：



（2）双要素情况：












]]></description><category></category></item><item><title><![CDATA[【代码实践】R语言，ugarchspec函数（待完善）]]></title><link>https://blog.youkuaiyun.com/Checkmate9949/article/details/120477034</link><guid>https://blog.youkuaiyun.com/Checkmate9949/article/details/120477034</guid><author>Checkmate9949</author><pubDate>Sat, 25 Sep 2021 21:16:51 +0800</pubDate><description><![CDATA[一、描述

创建单变量GARCH

二、用法


ugarchspec(variance.model = list(model = "sGARCH", garchOrder = c(1, 1), 
submodel = NULL, external.regressors = NULL, variance.targeting = FALSE), 
mean.model = list(armaOrder = c(1, 1), include.mean = TRUE, archm = FALSE, 
archpo]]></description><category></category></item></channel></rss>