Links table
Abstract and 1 introduction
2 methodology
3 results
3.1 Cryptopunks
3.2 NFT Market
4 discussion
5. Conclusion/ declarations/ references
extension
A.1 Implementation details
A.2 Detailed Information NFT and A.3 Map for Google Nft
2. Systematic
We described ways of analyzing gender and race at NFTS prices. First, we summarize our data collection (two sections 2.1 and 2.2) and then half how we define statistics between the gender and ethnic biases between the different NFTS (section 2.3). Steps are displayed in Figure 2. A more detailed description of the methods can be found and implementation can be found in the supplement.
2.1 Initial data collection
Our data collection consists of the processing NFTS on Opensea, which is the primary market for NFTS on ETAREUM. We inquire about the end of OpenSea “V1/Collcents” at the end of November 2022 [15] To recover NFT definition data and the last sale price. We choose 790 sets of Kagge Ethereum Nfts data collection [11] And NFTS of the top 30 days and OpenSea Loversover in around November 2021. After the data collection process, we end up 2.5 NFTS, each of which was dealt with.
2.2 Sexual signs and signs
Do not represent many NFT human groups, and therefore cannot be studied directly through the lens of sweat and sex. We choose groups that contain definition data with the phrase “male” and “female” and ends with a total of 44 of these groups with sexual stickers that represent different embodiment. Statistics can be found in these 44 groups with sex stickers in Table 6 in the A.2 appendix.
As much as we know, this is the first NFT data collection with sex stickers across many groups. However, there are often no race stickers in the descriptive data, so we are only limited to encrypted, Avastar, and dynamic combinations of groups with racing marks.
2.3 Statistical tools for prejudice analysis in sex and race
To determine the statistical importance of the hypothesis that NFTS is sold at less than male NFTS, we manage the unparalleled and unparalleled student student tests [22].
Un assistant test against T testing: For an unrestricted T test, we compare the average NFT selling prices for males for females. For the associated T test, we calculate the statistics t on the difference in male and male prices to the average daily price and the weekly price of NFT. The T test is used to isolate the difference between males against the price of females by repairing the price contrast over time.
Register transfer: Since the T-Test assumes the natural data, we apply the record conversion to process this. With NFTS is much more rare than common NFTS, NFT price distributions tend to follow the distribution of the Energy Law [14]. Inspired by how to follow the stock prices, a natural registration distribution [1]We applied the same transformation and found the distribution of the price record to be more natural. We point to the T test on the price record as a LOG T Test.
External pieces: Extreme values may occur due to the high selling prices for rare NFTS or very low selling prices due to human errors during inclusion. We treat extremist values using Winsorization [19]Or triming extremist values after a certain percentage. We report results of 0.1 %, 1 %, 2.5 % and 5 %.
The above described approach is also used to compare light and dark prices. We report results from groups of different T tests and methods of disclosure to show our conclusion, are still fixed regardless of the way we are testing statistical importance. For numbers and statistics in this paper, unless it is stated otherwise, we remove extremist values by 2.5 %.