Satellite observation has recently become very helpful for climate studies. Of particular interests for climate studies of extreme events is to provide a near-global coverage long-term fine resolution dataset for climate extreme analysis. In this study, PERSIANN (Precipitation Estimation from Remote Sensing Information using Artificial Neural Network) algorithm is used to estimate daily precipitation estimation of the past 30+ years based on longwave infrared imagery of GEO satellites. ISCCP (International Satellite Cloud Climatology Project) B1U dataset of NCDC is used for this activity. The constructed precipitation dataset is at 0.25 degree spatial and daily temporal resolution. Furthermore, the bias of PERSIANN daily estimation is further adjusted using monthly GPCP (Global Precipitation Climatology project) product (2.5 degree monthly data). Comparing with stage IV radar measurement in CONUS, GPCP bias adjusted PERSIANN precipitation estimation indicates improvement from non-adjusted PERSIANN estimation. The results show both storm detection and false alarms are improved after bias adjustment. The quasi-global daily PERSIANN precipitation data can be very useful for study climate extremes of tropical and extratropical regions (60S to 60N). Investigation of climate trends and extreme precipitation events using PERSIANN data is ongoing and will be reported in the meeting.