In both iOS and Android app Google Analytics show quite different numbers of unique users on chart and in the table:
The problem is the same even for different time intervals. A strange thing is that chart shows much more users than the table: an opposite case is not a problem, just several users can be mentioned in different events (or time interval, like in this question). But on the chart GA counts more people, than the sum of active users based on the events.
Is it a bug? Or GA just counts active users using some other stuff, not the events?
Here is my report settings:
Related
I created a time series type graph in grafana i attached the screenshot of same
This Graph Shows How many Applications for some specific time range (Last 7 days, Last 6 months etc.) is created in my Web Application.
Now i added a data link to another dashboard which shows time series graph of Applications with their current status.
http://localhost:3000/d/gr7ZaMank/loan-application-by-status?orgId=1&var-status=All&from=${__value.time}&to=${__to}
Now whenever i click a point on graph i am taking the time of that point in from parameter and in to parameter i am taking the last time.
As far as i know the graph shows data which is grouped by time.
Is there is any way to know how the date and time is grouped i break that date and pass it as a from and to in other dashboard for looking the acutal value of a point?
I recently completed upgrading bing maps from v7 to v8 for our dealerlocator. Since then we have observed increase in the no. of billed requests.
I did not change the way map API were being called during the upgrade process. Still, it shows an increase in no. of billed requests.
Can someone help me understand any specific reason for the same? Also how do I calculate no. of billed requests using JS in my code during development? Ideally I want to count the no. of requests for every load of this particular module. Please help :(
If you are simply loading a map a single billable transaction will be generated each time the map is loaded. If you load the map on page load then the number of page views would normally be around the same number as billable transactions. If you use the search or direction modules they will generate non-billable transactions as they are part of the map session.
If you use the auto suggest module, without a map, this will generate a billable transaction the first time a suggestion is selected.
We have done a lot of testing and have most V7 apps migrated to V8. We haven't seen any increases in usage in transactions in general. Any increases that were observed were due to the following:
App developers were comparing current month to previous month. This isn't accurate as web traffic generally fluctuates month to month. It is best to look at the last two years worth of reports to see if the month is question is typically a higher month and to also look at the average month on month growth as web traffic usually increases over time.
The new application attracted more users to use it. Sometimes additional marketing may have also been involved.
There was an issue with the code that the developer migrated which generated more transactions. For example, calling the map load function twice, once using the page onload like in V7 and once using the callback parameter in V8. This would result in twice as many map transactions being generated, but appear as only a single map on the page.
The previous application didn't use a key and either used v6 of Bing Maps or had CSS which hide the error message that appears in V7. We have seen this in a number of applications. Basically, the old application should have been generating more transactions than it was.
A list of the different ways transactions are generated is documented here: https://msdn.microsoft.com/en-us/library/ff859477.aspx
I'm trying to download 1 minute historical stock prices from Yahoo Finance, both for the current day and the previous ones.
Yahoo (just like Google) supports up to 15 days worth of data, using the following API query:
http://chartapi.finance.yahoo.com/instrument/1.0/AAPL/chartdata;type=quote;range=1d/csv
The thing is that data keeps on changing even when the markets are closed! Try refreshing every minute or so and some minute bars change, even from the beginning of the session.
Another interesting thing is that all of these queries return slightly different data for the same bars:
http://chartapi.finance.yahoo.com/instrument/2.0/AAPL/chartdata;type=quote;range=1d/csv
Replace the bold number with 100000 and it will still work but return slightly different data.
Does anyone understand this?
Is there a modern YQL query that can fetch historical minute data instead of this API?
Thanks!
Historical minute data is not as easily accessible as we all would like. I have found that the most affordable way to gather Intraday Stock Price data is to develop automated scripts that log price information for whenever the markets are open.
Similar to the Yahoo data URL that you shared, Bloomberg maintains 1-Day Intraday Price information in JSON format like this : https://www.bloomberg.com/markets/api/bulk-time-series/price/AAPL%3AUS?timeFrame=1_DAY
The URL convention appears easy to input on your own once you have a list of Ticker Symbols and an understanding of the consistent syntax.
To arrive at that URL initially though, without having any idea for guessing / reverse-engineering it, I simply went here https://www.bloomberg.com/quote/AAPL:US and used Developer Tools on my browser and tracked a background GET request which led me to that URL. I wouldn't be surprised if you could employ similar methods on other Price Data-related websites.
You can also write scripts to track price data as fast as your internet goes. One python package that I find pretty handy and is ystockquote
You can have it request price data every couple of seconds and log that into a daily time series database.
Yes there is other APIs.
I don't know if it can still help but if you need intraday data, there is a API on rapidapi called (Quotient) which allows to pull intraday (at 1-min level), EOD market (FX, Crypto, Stocks (US, CANADIAN, UK, AUSTRALIA, EUROPE), ETFs and Futures. It also provides earnings, dividends, splits and a lot others informations.
I`d like to get all events (from facebook) that will happen at a specific place.
for example:
Me, and two others people creating events in different days at a place X.
I need to get all info about the events created by me and others for that place.
I need to know how many users used my app. Now what is the correct definition of how many users used my app ?. What's the difference between active users, installed users and users who have used the app.
For example look at this graph i have attached. There are three numbers i am looking New installs (6554), Lifetime installed users(5431) and monthly active users(Highest at 17k+). What's the difference ? And what should be the answer when my question is how many users used my app ?
Here's an explanation of the different figures:
New installs - How many users have installed your app in a certain time period.
Lifetime Installed Users - How many users have ever installed your app (up to a certain point in time which is indicated when you hover on the ? beside the label).
Monthly/Daily/Weekly Active Users - How many users have actively used your app in a certain time period
It's your choice which of those figures you want to use to say how many 'use' your app, but DAU/WAU/MAU are probably the most honest in terms of real, actual users.