Message:

<p>[QUOTE=&quot;Roerbakmix, post: 4013992, member: 100731&quot;]Admittedly, there is a trend towards a decrease in value, however, please note that the number of coins in total is relatively small (n=318), as illustrated by the degree of uncertainty around the red line (the pinkish area. For those interested: the smoothed curve is a LOESS curve). It is difficult to statistically test this (this would include time-varying trends) - a more simple approach might be to compare averages of, say 2014 vs. 2018. </p><p><br /></p><p><br /></p><p>I agree: this would be a very good indicator. In essence, this could be compared to a &#039;case cross-over&quot; study design: the very same subject is &#039;exposed&#039; to the market, and yields a certain price. The benefit is, that all variables of the coin (grade, rarity, etc.) could be considered unchanged. The change in market price could thus be measured more clearly. </p><p>It is however largely impossible to do this automatically. </p><p><br /></p><p><br /></p><p>It is not, and this is a very good suggestion. I have just looked at inflation rates, but as someone with a limited understanding of inflation rates, I found it difficult to include this in my database. Preferably, I would need a datasheet with dates and inflation rates, so I could link historical prices to their inflation rates, and thus calculate a current estimated price. Any tips are welcome <img src="styles/default/xenforo/clear.png" class="mceSmilieSprite mceSmilie1" alt=":)" unselectable="on" unselectable="on" /></p><p><br /></p><p><br /></p><p>I believe SixBid includes this in their currency conversion. </p><p><br /></p><p><br /></p><p>Yes and no. Variance per se is not a problem if the data sample is sufficient enough. Suppose you want to model a prediction model (e.g. to predict a hammer price for a coin of your interest), variance is even a good thing (as it allows a model to differentiate better).[/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