Data for: Security Exchange Commission Forms K-10 filings - dataset of positive and negative words occurrence 1995-2008

All Forms 10-K analysis filled in the years 1995 to 2008. The original filing text files were mapped with “edgar” R package version 1.0.8. The filing text files were coded with the CIK number and date of Forms 10-K’s filing. Both a positive and negative dictionary were applied as suggested by Loughran and McDonald Financial Sentiment Dictionaries and as implemented in R edgar package We imposed a 10% frequency limit for both dictionaries .本数据集涵盖1995年至2008年间填报的全部10-K表格（Form 10-K）分析数据。原始申报文本文件通过R语言edgar包（edgar package）版本1.0.8完成映射处理；所有申报文本文件均已通过10-K表格的申报CIK编号（Central Index Key）与申报日期完成编码。本数据集采用Loughran与McDonald金融情感词典（Loughran and McDonald Financial Sentiment Dictionaries）中的正负向情感词典，并遵循R语言edgar包的实现逻辑进行调用；我们为两类情感词典均设置了10%的频率限制阈值。