DATA1001 Lecture Notes - Lecture 11: Standard Deviation, Kilogram, Measuring Instrument
20 Jul 2018
School
Department
Course
Professor
Document Summary
Here we simulate the weight of 20 trays of lamp chops from coles, using pricing from. 22/11/17. set. seed(1) chopweight = rnorm(20, 550, 5) chopweight. ## [1] 546. 8677 550. 9182 545. 8219 557. 9764 551. 6475 545. 8977 552. 4371. ## [8] 553. 6916 552. 8789 548. 4731 557. 5589 551. 9492 546. 8938 538. 9265. ## [15] 555. 6247 549. 7753 549. 9190 554. 7192 554. 1061 552. 9695 hist(chopweight) No matter how carefully any measurement is made, it could have turned out differently. Individual measurement = exact value + chance error. So if we weigh the same tray of lamp chops from coles 20 times, we would expect data like "chopweight". The best way to estimate the chance error is to replicate the measurement under the same conditions, and calculate the standard deviation. In any large enough series of careful replicated measurements, we expect to see a small percentage of extreme measurements, called outliers. One common convention is to define outliers as measurements more than 3 standard deviations away from the mean.
