批评Caged · 2010-10
“The paper claims their code and data is open, but the link is broken”
用推特公众情绪时间序列预测道琼斯指数涨跌,准确率 87.6%
论文检验从大规模推特流中得到的集体情绪是否与道琼斯工业平均指数(DJIA)相关。作者用 OpinionFinder(正负情绪)和 GPOMS(平静、警觉、确信、活力、友善、快乐六个维度)两种工具分析每日推文,再用格兰杰因果检验和自组织模糊神经网络考察情绪能否预测 DJIA 收盘价变化。结果称纳入某些特定情绪维度能显著提高预测准确率,预测每日涨跌的准确率为 87.6%,平均绝对百分比误差降低超过 6%。
评论者引用它说明推特情绪可用来预测市场,同时也有人指出其代码和数据链接已失效、方法论存在多处问题。
“The paper claims their code and data is open, but the link is broken”
“From my understanding, there are several problems with the methodology”
“There's a paper on it at Twitter mood predicts the stock market”
“its predictive ability has been examined several times academically within specific contexts (elections and markets - e.g. this one by Cornell Twitter mood predicts the stock market)”