Earnigns Reports (or simply Earnings) is an information concerning a company's earnings and its after-tax net income. This is a fairly complicated matter, and on TrendSpider, we use only a limited set of data points for this kind of data.
Using Earnings in scripting
Components supported: Scanner, Smart Checklist
You can use Earnings both "by date" and "by value".
This kind of data gives you a way for knowing things like "there's an Earnings report coming in 5 days" or "there was an earnings report exactly 5 days ago".
The latter, earnigns "by value", gives you access to numbers like
EPS (earnings per share),
Surprise% (EPS Surprise percent, which is a different between analysts' expectations for EPS and the actual EPS reported) and
Change% (which is a difference between the reported EPS and the EPS which has been reproted for the same quarter of a previous year).
NOTE ON OFFSETS: In scripting (i.e., scanning) Earnings Values work on a per-report basis (not a per-candle one). This means that "last" refers to the last report published, and "1 reports ago" refers to the previous report, disregarding of how long ago it had been published (in most cases it's 1 quarter).
NOTE ON "BY AT LEAST X%": "At least by X%" operator might behave couter-intuitive when one of your data points is zero. I.e., if your previous report's
Surprise% was 0 and for the last report it's 1, then the
Surprise% > Surprise% (1 report ago) at least by 10% script will be just as true as
Surprise% > Surprise% (1 report ago) at least by 1000%, because of the purely mathematical reasons (as the result of 1/0 is not defined and limits to ∞).