The following modified expression will flag the first occurence of the duplicated row as 1, assuming the table has an unique identifier and the data is ordered by it, such as the DimProduct table in AdventureWorksDW.
Index = CALCULATE(COUNTROWS(), ALLEXCEPT(Product, ‘Product'[ProductAlternateKey]), EARLIER(Product[ProductKey]) >= Product[ProductKey])

Here’s one solution. I’m sharing this, since this posting shows up in the top results of search engines.

For example, given this table “Host”:
ID HostName IpAddress CreatedDate
1 duplicatehost 192.168.1.2 2017-03-13
2 duplicatehost 192.168.1.2 2017-01-01
3 singlehost 192.168.1.3 2017-03-13

This DAX can be used to generate the following calculated table using “HostName” and “IpAddress” as the key for finding duplicates and getting the latest CreatedDate:

=FILTER(Host, [CreatedDate]=MAXX(FILTER(Host, [HostName]=EARLIER([HostName]) && [IpAddress]=EARLIER([IpAddress])), [CreatedDate]))

ID HostName IpAddress CreatedDate
1 duplicatehost 192.168.1.2 2017-03-13
3 singlehost 192.168.1.3 2017-03-13

There must not any duplication of the CreatedDate column for the key-grouping or duplicates will be present in the output. In my use case, duplicates are not possible.

Thanks for the reply. I’m using Analysis Services and connecting to my server live through Power BI Desktop. So, I cannot edit queries or use the Remove Duplicates feature in Power BI Desktop. Also, I need to create relationships between my tables, but this can only be done after the duplicates are removed, so I would rather do this in my Analysis Services model.

Hi Marty. Which tool are using: Power BI Desktop, Excel? As you mentioned, this blog is two-year old. Have you tried the Remove Duplicates feature in the Query Editor which is available in both Power BI Desktop and Excel?

In your example above, the row with ID 2 would be present in your new calculated table, but customers will wonder where the row with ID 1 is.