README.md

bloodstockR 1.0.0

R package with variety of datasets from Thoroughbred Bloodstock Sales, at the time of writing there are 6 datasets of Tattersalls sales from 2010 to 2015 (incomplete).

I recommend checking out the pinhooker R package which is along similar lines but contains a lot more data from other sales, and appears to be updated with more regularity.

Installation

Within R

devtools::install_github(repo = "durtal/bloodstockR")

Datasets

Only Tattersalls sales are in the package at the moment, hopefully others will be added in time (sales from Goffs etc.)

Tattersalls sales

Data was sourced from Tattersalls

Name | Sales included | Dim -----|----------------|----- tatts_2010 | dec_mares, dec_foals, dec_yearlings, autumn_hit, breeders_flat, oct_bk3, oct_bk2, oct_bk1, july, guineas, craven, feb | 7796 rows, 12 cols tatts_2011 | dec_mares, dec_foals, dec_yearlings, autumn_hit, breeders_flat, oct_bk3, oct_bk2, oct_bk1, july, breeze_up, craven, feb | 6946 rows, 12 cols tatts_2012 | dec_mares, dec_foals, dec_yearlings, autumn_hit, oct_bk3, oct_bk2, oct_bk1, july, guineas_hit, guineas, craven, feb | 7101 rows, 12 cols tatts_2013 | dec_mares, dec_foals, dec_yearlings, autumn_hit, oct_bk4, oct_bk3, oct_bk2, oct_bk1, july, guineas_hit, guineas, craven, feb | 7192 rows, 12 cols tatts_2014 | dec_mares, dec_foals, dec_yearlings, autumn_hit, oct_bk4, oct_bk3, oct_bk2, oct_bk1, july, guineas_hit, guineas, craven, feb | 7429 rows, 12 cols tatts_2015 | autumn_hit, oct_bk4, oct_bk3, oct_bk2, oct_bk1, july, guineas_hit, guineas, craven, feb | 5389 rows, 12 cols

The collect_tatts_sales function will collect data from a sale that is perhaps not included in the package datasets. The format Tattersalls presents data appears to be consistent across all sales, however no guarantees are made this function will perform as desired, if you discover an issue let me know.

Example
# load library
library(rBloodstock)
# load datasets, combine, and then remove individual datasets
data(tatts_2010, tatts_2011, tatts_2012, tatts_2013, tatts_2014, tatts_2015)
tattersalls <- rbind(tatts_2010, tatts_2011, tatts_2012, tatts_2013, tatts_2014, tatts_2015)
rm(tatts_2010, tatts_2011, tatts_2012, tatts_2013, tatts_2014, tatts_2015)

# load dplyr and ggplot
tattersalls %>%
    filter(grepl("sea the stars", sire)) %>%
    group_by(year) %>%
    ggplot(aes(x = factor(year), y = price)) +
        geom_boxplot(fill = "#d9220f") +
        RcappeR::theme_rcapper() + # my other package for a nice theme :-)
        labs(x = "Year", title = "Sea The Stars progeny @ Tattersalls")



durtal/bloodstockR documentation built on May 15, 2019, 6:01 p.m.