date_trunc('day',BLOCK_TIMESTAMP) as date,

SALE_TYPE,

count(distinct TX_GROUP_ID) as sale_no,

count(distinct PURCHASER) as buyer_no,

count (distinct nft_asset_id) as no_NFTs,

sum(TOTAL_SALES_AMOUNT_USD) as volume_usd,

avg(TOTAL_SALES_AMOUNT_USD) as average_volume,

min(TOTAL_SALES_AMOUNT_USD) as floor_price,

max(TOTAL_SALES_AMOUNT_USD) as highest_price,

median(TOTAL_SALES_AMOUNT_USD) as median_price,

avg(average_volume) OVER (partition by SALE_TYPE ORDER BY date ROWS BETWEEN 7 PRECEDING AND CURRENT ROW) as MA_7_Days,

sum(buyer_no) over (partition by SALE_TYPE order by date) as cum_buyers,

sum(sale_no) over (partition by SALE_TYPE order by date) as cum_sale,

sum(volume_usd) over (partition by SALE_TYPE order by date) as cum_volume

from algorand.nft.ez_nft_sales_fifa

where TOTAL_SALES_AMOUNT_USD>0

group by 1,2