Message:

<p>[QUOTE=&quot;Roerbakmix, post: 4010742, member: 100731&quot;]I&#039;m a bit late to the party, but have done some research in price development. Using data from sixbid for a few popular types (Caesar elephant denarius, Athens tetdrachm, and a few others), I tried to model a relation between predictors (e.g. estimate, grade, damage, slabbed vs unslabbed, etc.) and actual hammer price. I encountered various problems:</p><p>1) data is unstructured</p><p>2) sufficient amount of data is difficult to achieve (i.e. sixbid works by scrolling down. In order to obtain &gt;1000 datapoints, I installed an auto-scroller, which takes quite some time)</p><p><br /></p><p>Still, it&#039;s enjoyable. To overcome problem nr. 1, I programmed a text-mining algorithm to obtain information on</p><ul> <li>grade (grouped in four groups, ranging from fine to mint-state)</li> <li>slabbed vs unslabbed</li> <li>toned vs untoned</li> <li>damage (cracks, graffiti, cleaning marks, scrapes, etc.)</li> <li>banker marks</li> <li>off-center</li> <li>other notions (luster, brilliance, etc)</li> </ul><p>This topic is 1) on the price trend of high-end coins over time and 2) whether finding a (large) hoard influences the prices. These questions could of course be answered, but problem nr. 2 will probably make it a bit more difficult. Just for fun and giggles, I will try to demonstrate one example: the Julius Caesar elephant denarius.</p><p>Going back until 2012, I extracted 837 auction results:</p><p>[ATTACH=full]1055196[/ATTACH]</p><p><br /></p><p>As this topic is on high-end coins, I decided to take only the extremely fine and mint-state coins (n=391), remove the damaged, bankermarked, and off-centred coins, leaving a total of 318 coins with a substantially higher mean hammer price of €1177.</p><p><br /></p><p>Then, plotting the hammer price over time results in the following graph:</p><p>[ATTACH=full]1055197[/ATTACH]</p><p>which is fairly constant over time.</p><p><br /></p><p><br /></p><p>So, yeah, this is still a n=1 study. However, the script I&#039;ve coded is fairly accurate (it&#039;s coded in R) and could easily be extended to other coin types (at this moment, I&#039;m trying to answer question 2: is coin price affected by the finding of a hoard? using data of the Athens tetradracm). The most tedious part however is scraping obtaining data from Sixbid or other sites. So if anyone is interested enough, and would like information on a certain coin type, please pm me for more details <img src="styles/default/xenforo/clear.png" class="mceSmilieSprite mceSmilie1" alt=":)" unselectable="on" unselectable="on" />[/QUOTE]</p><p><br /></p>

Your name or email address:

Do you already have an account?

• No, create an account now.
• Yes, my password is:
• Forgot your password?

Stay logged in