G. Jogesh Babu is director of the Center for Astrostatistics at Penn State, and Eric Feigelson is the center's associate director and professor of astronomy and astrophysics at Penn State. The authors contributed this article toSpace.com's Expert Voices: Op-Ed & Insights <\/p>\n
After a century hiatus, astronomy and statistics recently reconnected, giving rise to the new field of astrostatistics. Some of today's most important issues in astronomy require sophisticated statistical modeling. NASA's Kepler mission has detected several thousand planets orbiting other stars, but it was through statistics that astronomers inferred that most stars have planetary systems and hundreds of millions of Earth-like planets probably exist in the galaxy. And in cosmology, statistics refined the parameters of the Lambda Cold Dark Matter (Lambda-CDM) consensus model of the universe, which suggests the universe expanded following a Big Bang 13.7 billion years ago, slowed by dark matter and accelerated by dark energy. <\/p>\n<\/p>\n
Insights from the ancients <\/p>\n
Such insights followed a long gap in the relationship between astrostatistics a term coined by us in our book of the same title published in 1996 and the broader field of astronomy. <\/p>\n
Astronomy is perhaps the oldest empirical science quantitative measurements of celestial phenomena were carried out by many ancient civilizations. The geometric models of the Platonists in ancient Greece proposed a cosmological model involving crystalline spheres spinning around a static Earth, a vision that endured in Europe for 15 centuries. <\/p>\n
It was another Greek natural philosopher, Hipparchus, who made one of the first applications of mathematical principles that we now consider to be in the realm of statistics. Finding scatter in Babylonian measurements of the length of a year, defined as the time between solstices, Hipparchus made the breakthrough decision to take guidance from the middle of a data range as the best value. <\/p>\n
Centuries later, a debate emerged about whether it is better to gather many data points or a few. On one side, the Arabic astronomer Ab Rayn al-Brn argued for more measurements to compensate for the dangers of propagating errors from inaccurate instruments and inattentive observers. In contrast, some medieval scholars advised against gathering repeated measurements, fearing that errors would compound rather than compensate for each other. It was in the 16th century that the utility of the mean to increase precision a favored method today was demonstrated with great success by Danish astronomer Tycho Brahe. <\/p>\n
In later centuries, some of the great thinkers of the day developed several elements of modern mathematical statistics specifically to address celestial mechanics, where Newton's Laws of Motion were producing astonishingly precise and self-consistent quantitative results for solar-system phenomena. <\/p>\n
In the late 18th century, in order to model cometary orbits, Adrien-Marie Legendre developed a system to fit noisy data to a mathematical model, which is now called the L2 least squares parameter estimation. The least-squares method became an instant success in European astronomy. <\/p>\n
<\/p>\n
The rest is here:<\/p>\n
Planet Hunting to Sky Surveys, Astronomy and Statistics Realign (Op-Ed)<\/a><\/p>\n","protected":false},"excerpt":{"rendered":" G. Jogesh Babu is director of the Center for Astrostatistics at Penn State, and Eric Feigelson is the center's associate director and professor of astronomy and astrophysics at Penn State. The authors contributed this article toSpace.com's Expert Voices: Op-Ed & Insights After a century hiatus, astronomy and statistics recently reconnected, giving rise to the new field of astrostatistics. Continue reading