科研速览 · Science Skim继续刷下去 · Keep skimming →
◇ Mendeley Data2026-08-03· Raw data

Dataset of article `Supporting and explaining stock investment decisions through Shapley Additive Explanations on LSTM and MLP-based agent models with user experience on one-decade backtesting’

Iván García‐Magariño, Javier Bravo

原始摘要(英文原文)· Original abstract
This dataset includes all the raw data of the article “Supporting and explaining stock investment decisions through Shapley Additive Explanations on LSTM and MLP-based agent models with user experience on one-decade backtesting “ submitted for publication consideration. The article includes both (a) the data generated from the simulations in our tool DI2XAI (a Dashboard for Intelligent Investment with eXplainable AI) with the corresponding statistical analyses, and (b) the data collected from the experiments with users. The dataset includes the following contents: - Raw data from backtesting simulations and their corresponding statistical analyses in the folder “RawDataBacktestingSimulationsAndAnalyses” (17 files inside) - Raw data from the replies of users to the questions after the experiments in folder “RawDataRepliesToQuestionnaires”. It includes the file "dashboardFormUserReplies_withCharts_anonymized.xlsx" with several tabs. The tab “estratificado_1” and “estratificado_2” contains the stratified data from tab “DashboardFormUserReplies_2026” tab. - Raw data from the profits of users in the backtesting user experiments compared to the Non-player Characters (NPC) using different strategies in the folder “RawDataUserExperiments”. The file "profitOverNpcs_withCharts_20260619_removed-duplicate-records_anonymized.xlsx" contains the results of user profits over NPCs, where the first tab contains all the results while the last tab contains the results separated in the different phases of the experiment.
读原文 · Read the paper ↗

AI 追问PRO

登录后使用 AI 追问

讨论区

登录后参与讨论

相关论文 · Related

Dataset of article `Supporting and explaining stock investment decisions through Shapley Additive Explanations on LSTM and MLP-based agent models with user experience on one-decade backtesting’ — 科研速览 Science Skim