Database: Uncovering investment management performance using SPIVA data

We used balanced panel data with 17 time dimensions from 2000 to 2016 and 14 cross sections (representing the different fund style/size categories) from the SPIVA US Scorecard, resulting in a total of 238 observations. The datasets of SPIVA scorecards only starts from 2000 as no data available for pre-2000 period. data is collected from the SPIVA US mid-year 2010 to the SPIVA US year-end 2016 reports, giving us semi-annual data from 12 SPIVA scorecards. The reason for using this time period is that the mid-year 2010 scorecard was the earliest one that is publicly available on the Standard and Poor’s website (Standard and Poor’s, 2017).本研究采用源自标普SPIVA美国记分卡（SPIVA US Scorecard）的平衡面板数据，该数据集包含2000年至2016年共17个时间维度，以及14个截面样本（对应不同的基金风格与规模类别），最终累计得到238条观测值。由于2000年之前无可用数据，SPIVA记分卡的数据集仅起始于2000年。本研究的数据采集自2010年中期至2016年末的标普SPIVA美国记分卡报告，由此获取了来自12份SPIVA记分卡的半年度数据。选取该研究时段的原因在于，2010年中期的记分卡是标普（Standard and Poor’s）官网可公开获取的最早一份SPIVA记分卡（Standard and Poor’s, 2017）。