M5 EDA(1)

    技术2022-07-11  93

    (codes见Kaggle kernel-notebook) 各州销量占比 CA 43.6%;TX 28.8%; WI 27.6%; 各州各商店占比 CA_1,26.8%; CA_2,19.8%; CA_3 39.0%; CA_4 14.4%; 0 26.845926 Name: total_sales, dtype: float64 1 19.82691 Name: total_sales, dtype: float64 2 39.016448 Name: total_sales, dtype: float64

    TX_1 2 3 4 29.606277 Name: total_sales, dtype: float64 5 38.173352 Name: total_sales, dtype: float64 6 32.220372 Name: total_sales, dtype: float64

    WI_1 2 3 7 28.415115 Name: total_sales, dtype: float64 8 36.113151 Name: total_sales, dtype: float64 9 35.471735 Name: total_sales, dtype: float64

    类别销售占比 FOODS 68.6%; HOUSEHOLDS 22.0%; HOBBIES 9.3%

    部门销量占比 dept_id total_sales 0 FOODS_1 7.744896 1 FOODS_2 11.613935 2 FOODS_3 49.276010 3 HOBBIES_1 8.518799 4 HOBBIES_2 0.804227 5 HOUSEHOLD_1 17.505829 6 HOUSEHOLD_2 4.536305

    FOODS2 3 占FOOD比例 HOUSEHOLDS1占HOUSEHOLDS比例 1 16.92134 Name: total_sales, dtype: float64 2 71.794455 Name: total_sales, dtype: float64 5 79.419847 Name: total_sales, dtype: float64

    各类商品在各州的销量占比 total_sales_CA total_sales_TX total_sales_WI cat_id FOODS 42.573360 28.734616 28.692024 HOBBIES 49.632364 25.998220 24.369416 HOUSEHOLD 44.468674 30.041745 25.489580

    FOODS和HOUSEHOLD方面40%的销售量来自California,而HOBBIES销售量50%。 HOBBIES和HOUSEHOLD方面25%的销售量来自Winscoin,FOODS销售量29%来自Winscoin。 FOODS和HOUSEHOLDS近30%来自Texas,而业余爱好用品销售额25%来自Texas。

    total_sales state_id CA TX WI cat_id FOODS 0.669431 0.685561 0.713941 HOBBIES 0.106010 0.084255 0.082368 HOUSEHOLD 0.224559 0.230184 0.203691

    Winscoin在FOODS上花费最高72%,在HOBBIES上花费8%,在HOUSEHOLD中花费20%。 California在HOBBIES所有花费是所有州最高的11%,花费67%在FOODS上,,22%在HOUSEHOLD上。 Texas在HOUSEHOLD花费最高23%,FOODS上68%,HOBBIES上,在所有州中花费为8%

    每家店各类商品的销售额 total_sales dept_id FOODS_1 FOODS_2 FOODS_3 HOBBIES_1 HOBBIES_2 HOUSEHOLD_1 HOUSEHOLD_2 store_id CA_1 7.376371 11.498041 51.022666 10.671615 0.716452 14.291168 4.423687 CA_2 12.623167 5.852088 42.737907 10.171235 1.046878 19.468066 8.100660 CA_3 6.586076 10.748540 49.841547 7.914764 0.674185 19.190548 5.044341 CA_4 8.371275 13.093504 47.240084 13.204210 0.550653 12.937790 4.602483 TX_1 6.535066 11.286507 49.703215 6.899086 0.769576 20.478574 4.327978 TX_2 7.048641 10.831999 51.634984 7.771571 1.044109 17.384977 4.283720 TX_3 6.757065 12.385139 49.224726 7.553655 1.105064 19.053607 3.920743 WI_1 9.840666 11.668028 45.267585 11.810365 0.923916 15.981066 4.508375 WI_2 8.113585 16.696776 48.052953 5.019734 0.637560 18.562283 2.917110 WI_3 6.204706 12.624853 54.767834 6.627481 0.632691 15.831682 3.310753 WI-3销售最不均衡,54%的销售来自FOODS-3,7%的销售来自HOBBIES。 CA-2销售最为均衡,与其它商店相比,FOODS-3销售量最低,而FOODS-1和HOUSE-HOLD2销售量最高。 对于所有商店而言,FOODS-3、HOBBIES1和HOUSEHOLD-1销售占比最高

    Price Data item_id store_id sell_price_max sell_price_min price_change category 0 HOUSEHOLD_2_406 WI_3 107.32 3.26 104.06 HOUSEHOLD 1 HOUSEHOLD_2_406 WI_2 61.46 12.46 49.00 HOUSEHOLD 2 HOUSEHOLD_2_466 TX_1 52.62 6.46 46.16 HOUSEHOLD 3 HOUSEHOLD_2_178 TX_1 44.36 3.00 41.36 HOUSEHOLD 4 HOUSEHOLD_2_250 WI_2 34.18 3.36 30.82 HOUSEHOLD HOUSEHOLD产品价格变动最为明显,尤其是HOUSEHOLD_2产品在WI州中变化最大。 WI_3商店的HOUSEHOLD_2_406价格变化最大,从最低价格的3.26美元上涨了32倍,达到了107美元。

    产品FOODS_1_014的价格在若干年未发生变动.

    可以得知信息: 大多数产品价格并未发生改变,大多数产品价格波动范围在10-15美元 其中HOUSEHOLD的产品的价格波动最为明显

    国家 宗教 文化 体育类节日各包括哪些具体节日 Cultural National Religious Sporting 0 ValentinesDay PresidentsDay LentStart SuperBowl 1 StPatricksDay MemorialDay LentWeek2 NBAFinalsStart 2 Cinco De Mayo IndependenceDay Purim End NBAFinalsEnd 3 Mother’s day LaborDay OrthodoxEaster NaN 4 Father’s day ColumbusDay Pesach End NaN 5 Halloween VeteransDay Ramadan starts NaN 6 Easter Thanksgiving Eid al-Fitr NaN 7 NaN Christmas EidAlAdha NaN 8 NaN NewYear Chanukah End NaN 9 NaN MartinLutherKingDay OrthodoxChristmas NaN 一年有10个国家和宗教活动,6个文化活动和3个体育活动,总共30个活动。

    每个州每年每月都有10个snap_day 举办月份大致相同 year month snap_CA snap_TX snap_WI total_events 33 2013 10 10 10 10 30 49 2015 2 10 10 10 30 35 2013 12 10 10 10 30 36 2014 1 10 10 10 30 37 2014 2 10 10 10 30 … … … … … … … 27 2013 4 10 10 10 30 28 2013 5 10 10 10 30 29 2013 6 10 10 10 30 30 2013 7 10 10 10 30 0 2011 1 0 0 0 0

    日总销售额时间走势 总的趋势是日销售额在增长。

    三个州的月度销售时间序列 CA州的销售额一直是最高的。与其他州相比,CA州8月份的高峰最为明显。季节性对CA州销售的影响最大 WI州的销售额增长率是历年来最高的。2013年之前,WI州的销售额低于TX州。2013年至2015年8月,TX州和WI州的销售额相似。2015年8月之后,它的销售额高于TX州。看看是什么导致WI在过去几年中大幅增长。

    Processed: 0.010, SQL: 9