In conclusion, our study answers the question that there exist some similarities of stocks among different sectors and makes a small progress. Back to the question in the introduction, from the results we can see that consumer defensive/discretionary and technology both have 'A' level volume and 'small' level risk. However, our study has some drawbacks. First, although our datasets have 14 attributes and over 5000 records at the beginning, they contain lots of missing values, invalid data and outliers. Therefore, after cleaning (drop or fill using statistical number) and merging, our dataset only contains over 2000 records. We think the number of records is not enough to give a pretty valid result. Second, there are only 5 numerical variables in our dataset and we selected three to do the clustering analysis. We think the number of attributes is not enough to give a pretty valid result. If we include more data records and more attributes such as net income, total revenue of each stock, clustering analysis may be more accurate. Finally, our data only contains information about a certain point of time, rather than a period of time. As a result, we cannot grab the change of the pattern over time. We believe that the time-period data can give more information about the pattern and trend of stocks.